ti)
i ~
I
I ·~ 10-<
:
---t
iQ I
.......
U
ctl
i!
!
I; '::U !
~
I ......
!
! ~
1 0
I ......
18 "
, ...... i S:: 1 0
1
l U
I "'O
S i ~ v l ....c: s::
tI) ! O
+-> ,
~ i! . ~
~ ~
I
I
O~ ,~e
~I;>
6'< 1" n " 1,-) L58 1 v
-< o ~(
Werner Leonhard
Contrai of Electrical Drives Third Edition F~J= ~~D1,~
U.F.M.G.· BIBLIOTECA UNIVERSITÁRIA
rll rlllfl lrlllrl~lllllllllllr I I
26780505
NAo DANIFIQUE ESTA ETlQUETA With 299 Figures
SI lrillgcr /l a/ iII l/rldt"!{Jag Nnl' Yo rJe
Hilrct:!ollll [[oJlg
KOllg
I./JJJd(}}/
Mi/tl IIO
f'lI! i·o I
.'
.
li.ll",II11'l'r ing
I
1ONIINI LIIIIII\RY
~ (n.l:(llllll t'
1: ,/11 1[ '
III
I) 1.//WWW."1II II 1/ II'I,dl l/" 1\11 I! II ,/
' i
~
, .,
t>
~ . ~ Spri ngcr t~~ •.'"
~
_ ._._-- -
..
_ -~
BIBLIU 1 1' C "
I q/" llI .iI. n..
I , N/II III1 IAIJ IA
ill~.
Preface
Professor WERNER LEONHARD
Tl'chnische Universitãt Braunschweig IlIsti tut für Regelungstechnik II a ns Sommer Str. 66 ,\1) 106 Braunschweig I' II/(./il:
[email protected] 1/llcrnet: http://www.ifr.ing.tu-bs.de
'JÇLtb 14 ,~'~ 1
1 \::1 ,'~{~t.
6',l,: '.p.,~' ,:'; :!
':: . .
r ~:: F' ':~ (\
':.' ~~~ c'::; . !.~~ .
IS I\N \
:,~()-41820-2 Springer-Verlag Bedin Heidelberg NewYork
I 01" ,II Y "I ( :0111'.1'<:" Cataloging-in-Publication Data 1 "01111111.1. W(' rncr:
1 ",,\, 01 01 l' lc. (rical Drives. - Third edition.1 Werner Leonhard. - Berlin; Heidelberg; New York;
11 ,", ,,111,0:1; Iln"I', Kong; London; Milan; Paris; Singapore; Tokyo: Springer 2001
(IIIIV'I,I\'['
I:;III~
SYSlt'IIIS)
I ','lO 111\20 .. 2
TI ... : worl,
i, slIbjcct to copyright. All rights are reserved, whether the whole Dr part ofthe material is ,pccifically the rights Df translation, reprintíng, reuse Df illustrations, recitation, I" " " .I. "s\ illg , reproduction on microfilm or in otherways, and storage in data banks. Duplication of I 100',p""li ..,,\ ion or parts thereofis permitted only under the provisions ofthe German Copyríght Law "I :;I'\,\('m"er 9, 1965, in its current version, and permission for use must always be obtained from :'1" illl',"r Vcrlag. Violatíons are liable for prosecutíon act under German Copyright Law.
, "'I< .. roo('d,
:qll '001',<'1' Verlag is a company in the BertelsmannSpringer publishing group ,,\ I I" Iwww.springer.de
~, ' S" r;III',('r , V crlag Berlin Heidelberg New York 2001
l' rI "I",1 iII (;cnnany
'1'\". 00 SI' "fl',ellcral dcscriptive names, registered names, trademarks,etc. in this publication does not ioo'ply, I'vcn iII lhe "bsencc or a spccific statement, that ,uch names are excmpl from the relevant "r"I,.•'liv(' law, a"d regulat;o" s llnd tht:rdore freI: for I',cllnal US(' . 'L'yp rtJdltnH: ( :111111'1 ~t 1~';lIly ( '''V I' '
by
Illlllltlr
I h' NII",, : .i<- \olil'" 11",1;"
l'I " ~1c'dtlll ll . h Lfl"' ~ I"'111'1
.'~I'IN : IOH\\II\)C\
(1 _~ /\O .loJl.II
'1.1 \.' 10
Electrical drives play an important role as electromechanical energy convert ers in transportation, material handling and most production processes. The ease of controlling electrical drives is an important aspect for meeting the in creasing demands by the user with respect to flexibility and precision, caused by technological progress in industry as well as the need for energy conser vation. At the sarne time, the control of electrical drives has provided strong incentives to control engineering in general, leading to the development of new control structures and their introduction to other areas of controI. This is due to the stringent operating conditions and widely varying specifications - a drive may alternately require control of torque, acceleration, speed or position - and the fact that most electric drives have - in contrast to chem ical or thermal processes - well defined structures and consistent dynamic characteristics. During the last years the field of controlled electrical drives has undergone rapid expansion due mainly to the advances of semiconductors in the form of power electronics as well as analogue and digital signal electronics, eventu ally culminating in microelectronics and microprocessors. The introduction of electronically switched solid-state power converters has renewed the search for adjustable speed AC motor drives, not subject to the limitations of the mechanical commutator of DC drives which dominated the field for a century. This has created new and difficult control problems since the mechanically simpler AC machine is a much more involved control plant than a DC motor; on the other hand, the fast response of electronic power switching devices and their limited overload capacity have made the inclusion of protective control functions essentiaI. The present phase of evolution is likely to continue for some years, a new steady-state is not yet in sight. This book, originally published 1974 in German as "Regelung in der elek trischen Antriebstechnik", was an outcome of lectures the author held for a number of years at the Technical University Braunschweig. ln its updated En glish version it characterises the present state of the art without laying claim to complete coverage of the field. Many interesting details had to be omitted, which is not necessarily a disadvantage since details are often bound for early obs()les(:(~Ilc(~ . ln selecting and presenting the material, didactic view points ha.vc also hC(~1I clllIsidcrpr]. A prerequisite for the reader is a basic knowledge
VI
Preface
Preface
"r !l 0wer electronics, electrical machines and control engineering, as taught in 1111d, I1Ilcler-g r aduate electrical engineering courses; for additional facts, re (11IllSC is made to specialliterature. However, the text should be sufficiently ::,,11' coutained to be useful also for non-experts wishing to extend or refresh Ilwir kuowledge of controlled electrical drives. Tbcy consist of several parts, the electrical machine, the power converter, !.II(' ('oIltrol equipment and the mechanicalload, all of which are dealt with in v:uyiJlg depths. A brief resume of mechanics and of thermal effects in electri ()II IIlachines is presented in Chaps, 1-4 which would be skipped by the more 1 ':: I'I ~ri(~llCed reáâer. Chaps. 5-9 deal with DC drives which have for ove r a "'III,lIry been the standard solution when adjustable speed was required. This p;\.rl. or the text also contains an introduction to line-commutated converters ,I:: IISI:
l'1';iII Jlillg with a general dynamic model of the symmetrical AC motor, valid iII IIpC" the steady-state and transient condition. This is followed in Chapo 11 II\' ali overview of static converters to be employed for AC drives. The con 1.1'01 ;1.spects are discussed in Chaps. 12-14 with emphasis on high dynarnic l'I'rI'Ol'lllanCe drives, where microprocessors prove invaluable in disentangling LIli' 1I1111Civariate interactions present in AC machines. Chapter 15 finally de ::nil)('s some of the problems connected with the industrial application of .IriVl's , 'l'ltis cannot by any means cover the wide field of special situations Wil.lt wllich the designer is confronted in practice but some frequently en '''IIIII.I'II'd katures of drive system applications are explained there. It will 111''''"111' ~"Ifliciently clear that the design of a controlled drive, in particular "I. Iilrl','" pow(~r ratings, cannot stop at the motor shaft but often entails an ;(11:0.11':" :: ,.I' L"(~ whole electro-mechanical system. III vinv 01' Lhe fact that this book is an adaptation and extension of an "111 dical.i, 'll-oricIltated text in another language, there are inevitably problems 1.' 1111 1" 'I ';: \.rd 1.0 symbols, the drawing of circuit diagrams etc. After thorough ","'::lIll.a.t.iolls with competent advisors and the publisher, a compromise so I"Li!)1I was adopted, using symbols recommended by IEE wherever possible, 1.111. rdailliug the authors usage where confusion could otherwise arise with I,i:: r(';ukrs at home. A list of the symbols is compiled following the table of ,'IIIII,I'III.S, The underlying principIe employed is that time varying quantities ;),1'" IISll1\.lly dcnoted by lower case latin letters, while capitalletters are ap pli,'" 1.11 parameters, average quantities, phasors etc; greek letters are used pt'l'dolllÍllalltly for angles, angular frequencies etc. A certain amount of over 1:'1 1 iH Illl
VII
The author wishes to express his sincere gratitude to two British col leagues, R. M. Davis, formerly of Nottingham University and S. R. Bowes of the University of Bristol who have given help and encouragement to start the work of updating and translating the original German text and who have spent considerable time and effort in reviewing and improving the ini tial rough translation; without their assistance the work could not have been completed. Anyone who has undertaken the task of smoothing the transla tion of a foreign text can appreciate how tedious and time-consuming this can be. Thanks are also due to the editors of this Springer Series, Prof. J. G. Kassakian and Prof. D. H. Naunin, and the publisher for their cooperation and continued encouragement. Braunschweig, October 1984
Werner Leonhard
Preface to the 2 nd edition During the past 10 years the book on Control of Electrical Drives has found its way onto many desks in industry and universities all over the world, as the author has noticed on numerous occasions. After a reprinting in 1990 and 1992, where errors had been corrected and a few technical updates made, the book is now appearing in a second revised edition, again with the aim of offer ing to the readers perhaps not the latest little details but an updated general view at the field of controlled electrical drives, which are maintaining and extending their position as the most ftexible source of controlled mechanical energy. The bibliography has been considerably extended but in view of the con tinuous strearn of high quality publications, particularly in the field of con trolled AC drives, the list is still far from complete. As those familiar with word processing will recognise, the text and figures are now produced as a data set on the computer. This would not have been possible without the ex pert help by Dipl.-Ing. Hendrik Klaassen, Dipl.-Math. Petra Heinrich, as well as Dr.-Ing. Rüdiger Reichow, Dipl.-Ing. Marcus Heller, Mrs. Jutta Stich and Mr. Stefan Brix, to whom the author wishes to express his sincere gratitude. Peter Wilson has been reading some of the chapters, offering valuable sugges tions, The final layout remained the task of the publishers, whose patience and helpful cooperation is gratefully appreciated. Braunschweig, May 1996
Werner Leonhard
Prefacc to the 3 rd Edition '1'1111. (,;\.\'Iy c\"l'ldio" of LIli: pllblislllll's :.;Locks oll('I'<,d Hll opportullit,y of again II 1''':!.I,i "I': I.ltc' IIClClII , IIlakil'l', IIliJlOI' 1'01'l'I'cl.i'III:; :utel I'lllal'l':illJf, SOIlIt' SlIh.i,'i'I,H I
V III
Preface
whilc t.hc main part of the book was left unchanged. Tackling this work 1(''1uircd the encouragement by good friends but the author would still have '(:11 lIuable to realise it alone. He expresses his sincere gratitude to the helpful \(':;('archers at the Institut für Regelungstechnik, particular1y Dipl.-Ing. Jan I \\ ... kcr and Dipl.-Ing. Frithjof Tobaben who were always ready to resolve ::Illl.wilrc-related crises on the computer, as well as Dipl.-Ing. Klaus Jaschke ;1.lId C;\llél. Wirtsch.-Inform. Danny Wallschlager, our definitive ~TEX-experts wliosc participation was essential for undertaking the task. Dr.-Ing. Sõnke 1\ ()ck suggested a clearer definition of the magnetic leakage which had gone II II lIot.iccd beforej finally, I want to thank Prof. Walter Schumacher, now head (Ir Ui(' laboratory, for his continued interest. Werner Leonhard 1\ r:tllllscltweig, Spring 2001
Contents
1)(
Introduction ................................................. .
1
1. Elementary PrincipIes of Mechanics ........ . ............ .
1.1 1.2 1.3 1.4 1.5
7
Newtons Law ......................................... . 7
Moment of Inertia ............. .. ...................... . 9
Effect of Gearing . ..... .......... . .............. ... .... . 11
Power and Energy .. .. .......................... ...... . . 12
Experimental Determination of Inertia ................... . 14
2. Dynamics of a Mechanical Drive ......................... . 2.1 Equations Describing the Motion of a Drive with Lumped
Inertia .......................................... . .... . 2.2 Two Axes Drive in Polar Coordinates . .... .......... . .... . 2.3 Steady State Characteristics of Motors and Loads ......... . 2.4 Stable and Unstable Operating Points .................... .
17
3. Integration of the Simplified Equation of Motion ........ . 3.1 Solution of the Linearised Equation ............. . ........ . 3.1.1 Start of a Motor with Shunt-type Characteristic at
No-Ioad ........................................ . 3.1.2 Starting the Motor with a Load Torque Proportional
to Speed ........................ . .............. . 3.1.3 Loading Transient of the Motor
Initially Running at No-Ioad Speed .. . .. . .......... . 3.1.4 Starting of a DC Motor
by Sequentially Short-circuiting Starting Resistors ... . 3.2 Analytical Solution of Nonlinear Differential Equation ...... . 3.3 Numerical and Graphical Integration ..................... .
29
29
4. Thermal Effects in Electrical Machines .................. . 4.1 Power Losses and Temperature Restrictions ............... . 4.2 Hcating of a Homogeneous Body ....... .. ....... . ..... . . . 1.:~ Difrer('lll: Morles of Operation ...... . . .............. ... .. . '1.:\.1 (;()ilt.illIlOIlS Dllty ..... ' ..... . . .. . . ...... . . . ...... .
17
20
22
26
30
32
33
35
38
39
43
43
44
48
1K
x
Contents
Contents
4.3.2 4.3.3 ~ ',.
O.
"(
Short Time Intermittent Duty ..................... 48
Periodic intermittent duty . . . . . . . . . . . . . . . . . . . . . . . .. 49
SeparateIy Excited DC Machine . . . . . . . . . .. . . .. . . . . . . . . . .. 5.1 Introduction .................... ·.····················· G.2 Mathematical Model of the DC Machine ..... ..... ........ G.3 Steady State Characteristics with Armature and Field Control 5.3.1 Armature Control . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . .. 5.3.2 Field Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5.3.3 Combined Armature and Field Controlo . . . . . . . . . . . .. !:iA Dynamic Behaviour of DC Motor with Constant Flux .. . . . ..
51
51
54
56
57
58
61
64
DC Motor with Series Field Winding . . . . . . . . . . . . . . . . . . . .. 69 0.1 Block Diagram of a Series-wound Motor. . . . . . . . . . . . . . . . . .. 70
G.2 Steady State Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
. ControI of a Separately Excited DC Machine .. . . . . . . . . . .. 77
7.1 Introduction .................... ·······················
7.2 7.:3 'IA K.
77
Cascade Control of DC Motor in the Armature Control Region 79
Cascade Control of DC Motor in the Field-weakening Region 90
Supplying aDC Motor from a Rotating Generator.. . . . . . . .. 93
Static Converter as a Power Actuator for DC Drives ..... 97 ~u Electronic Switching Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 97
H.2 Line-commutated Converter in Single-phase Bridge Connection102
~u Line-commutated Converter in Three-phase Bridge Connection119
H.tJ Line-commutated Converters with Reduced Reactive Power .. 130
H.:) Control Loop Containing an Electronic Power Converter .... 133
n. (;oulrol of Converter-supplied DC Drives ................. 139
~u DC Drive with Line-commutated Converter ........ . ....... 139
~).2 DC Drives with Force-commutated Converters ............. 148
10. Syrnrnetrical Three-Phase AC Machines .. , ............... 163
10.1 Mathematical Model of a General AC Machine .......... ·.· 164
10.2 Induction Motor with Sinusoidal Symmetrical Voltages in
Steady State ........ .. ............ . . . .................. 176
10.2.1 Stator Current, Current Locus .................. ... 176
10.2.2 Steady State Torque, Efficiency .................. . . 182
10.2.3 Comparison with Practical Motor Designs .......... , 186
10.204 St.art.ing of lhe Induction Molor ......... .. .... .... . 187
.Ill. :.\ Indlld.ioll Mot.()r with lmpresscd Volt.ar;c·s or t\ I'hit.r,\ry Wave-
rOI'lIl :l . .
' .... . . . . ........ ,
••• . . • , .. . . , •• . . . . . . .
10/1 111.11\1'1.1011 M,J\,,!' wit.h lJn:iyllllllC't.1 i,';" 1,11 ... VIIII,il.l',C·: i iII S t.c ·;\.dy ~; 111.1."
"
. . •• , • , •••
190
:J.m
1004.1 10.4.2 10.4.3 10.4.4
Symmetrical Components ......................... Single-phase Induction Motor ........ ..... .... .. ... Single-phase Electric Brake for AC Crane-Drives ..... Unsymmetrical Starting Circuit for Induction Motor ..
XI
202
206
209
211
11. Power Supplies for Adjustable Speed AC Drives . ... ...... 215
11.1 Pulse width modulated (PWM) Voltage Source Transistor
Converter (IGBT) .................... . ................. 11.2 Volt age Source PWM Thyristor Converter ................. 11.3 Current Source Thyristor Converters ....... ........ ..... .. 11.4 Converter Without DC Link (Cycloconverter) ..............
218
225
232
236
12. Control of Induction Motor Drives ....................... 241
12.1 Control of Induction Motor Based on Steady State Machine
Model ................................................. 242
12.2 Rotor Flux Orientated Control of Current-fed Induction Motor252 12.2.1 PrincipIe of Field Orientation .......... .. .......... 252
12.2.2 Acquisition of Flux Signals .. ..... ......... . .. . .... 260
12.2.3 Effects of Residual Lag of the Current Control Loops . 262
12.2.4 Digital Signal Processing ... .. ..................... 265
12.2.5 Experimental Results ............................. 268
12.2.6 Effects of a Detuned Flux Model ...... . . . .......... 269
12.3 Control of Voltage-fed Induction Motor ................... 275
12.4 Field Orientated Control of Induction Motor with a Current
Source Converter ......... . .......... ...... ............. 281
12.5 Control of an Induction Motor Without a Mechanical Sensor. 289
12.5.1 Machine Model in Stator Flux Coordinates .......... 289
12.5.2 Example of an "Encoderless Control" ... ... ......... 291
12.5.3 Simulation and Experimental Results ............... 296
12.6 Control of an Induction Motor Using a Combined Flux Model 298
13. Induction Motor Drive with Reduced Speed Range ....... 303
13.1 Doubly-fed Induction Machine with Constant Stator Fre quency and Field-orientated Rotor Current ................ 303
13.2 Control of a Line-side Voltage Source Converter
as a Reactive Power Compensator ...................... .. 317
13.3 Wound-Rotor Induction with Slip-Power Recovery .......... 323
14. Variable Frequency Synchronous Motor Drives ........... 329
14.1 Control of Synchronous Motors with PM Excitation .. . . .... 331
14.2 Synchronous Motor with Field- and Damper-Windings ..... . 342
14.3 Synchronous Motor with Load-commutated Inverter (LCI-
Driv(') ........ . .......... .... .......... ..... ......... . 349
X II
Contents
I !:i . Some Applications of Controlled Electrical Drives ........ 15.1 Speed Controlled Drives ................................. 15.2 Linear Position Control ................................. 15.3 Linear Position Control with Moving Reference Point ....... 15.4 Time-optimal Position Control with Fixed Reference Point .. 15.5 Time-optimal Position Control with Moving Reference Point .
363
364
373
383
389 396
Abbreviations and Symbols
Bibliography .................................................. 402
IIHlcx .... , ........... , ........................................ 455
1 Equations ln all equations comprising physical variables, they are described by the prod uct of a unit and a dimensionless number, which depends on the choice of the unit. Some variables are nondimensional due to their nature or because of nor malisation (p.u.). 2 Characterisation by Style of Writing
i(t), u(t), etc.
1:, I d , u, Ud , etc. I, U, etc. U, etc. i.(t), :!!(t), etc.
L
f (t), :!!* (t), I*, U', etc. li(t), l:!!(t), etc. 1(8) = L(i(t)) etc.
instantaneous values average values RMS-values complex phasors for sinusoidal variables complex time-variable vectors, used with multiphase systems conjugate complex vectors or phasors vectors in special coordinates Laplace transforms
3 Symbols
Abbreviation
Variable
Unit
a(t)
current distribution linear acceleration nondimensional factor area nondimensional field factor magnetic flux density clectrical capacity theonal storaw' capnci t.y
A/m m/s2
A b D C "
f)
dlLl II pill r~ r a Lio
m2 T = Vs/m 2 F = As/V J;oC = Ws;oC
,(/), I ~, E
r ,,' (.'i ) 'I 'I (I)
"
(
ii ,(1) , I , ,/
I
I,;
, " /I
1/1,( /) II/
'/I
N
I' ( /) , I' () /'
I,'
" :1 \
I J
W
"
," '
/ "
'
/I
(I), II,
" (I) \I
/1/(1 ) ,/'
:'1 ('.'.""
\' /. /I'
Abbreviations and Symbols
A bbreviations and Symbols
'\IV
II
induced voltage, e,m.f, , frequency force transfer function gravitational constant unit impulse response weight gain airgap current inertia nondimensional factor torsional stiffness length inductance torque mass mutual inductance speed, rev /min
number of turns
power
reactive power
radius
resistance
Laplace variable distance
slip
time
time constant
voltage
velo city
unit ramp response
volume
unit step response
energy
control variable
actuating variable
disturbance variable discrete Laplace variable admitt.a.nce illlj)('dallce codlicicut,
or II(';\'/.
t,rnll :';I'('I'
lil'ill/', ;\.11/,:1(' 1111['," LII '
,1.1 '''' '1''I'a/,IIII'
v Hz = l/s N m/s2 N m A kg m 2 Nm/rad m H = Vs/A Nm kg H l/min
I
I
i
1\ ,I
!' I
W VA m
n
I
0:, {3, 15, (, ç, >., /l, {2 etc. angular coordinates "1= 21f/3 15 load angle L1 difference operator angle of rotation ê efficiency ." {) temperature absolute temperature magnetomotive force, m.m.r. e coefficient of permeability /lo v integer number (J' leakage factor normalised time, angle T = J w dt, wt phase shift
ia ie UF
s
iSd, ZSq
V
m/s
ZS
iR iRd, iRq
Zm i mR ZmS
m3 J =Ws
1/f2 = S .rl W /1I1;~ "(;
rad rad rad
oe K A H/m
rad rad Wb = Vs Wb = Vs rad/s
41ndices
rad/s m
S
xv
mL mM mp Sp
armature current exciting current field voltage stator current rotor current direct and quadrature components of stator current direct and quadrature components of rotor current magnetising current magnetising current representing rotor flux magnetising current representing stator flux load torque motor torque pull-out torque pull-out slip
xV I
Abbreviations and Symbols
Abbreviations and Symbols
Y1 ----171
., (;('aphical Symbols
4 kí~
T~~=y
_~
v
~
T
-
4l~::í~
dx
~
~k~
= G (T !EL ~
+ y)
T~~+x=Gy
PI controller
first order lag
~8t!j~
~~I~) ~ ...............
Tl~~+X=G(T2~+Y)
x
41 t--:
YI
x(t) = G y (t - T)
delay
= Yt!Y2
division
x
= f(y)
nonlinearity
x
=y
= x =
X
Xmin X max
for Xmin < y < for y ~ Xmin for y ~ X max
X max
limiter
x = Yl - Y2
summing point
x = Yl Y2
multiplication
•
~i lu
X
..
.
---+I-~r---
X
--_.~
x
first order lead/lag
-------..~
~O
•
xmin
G r;; r,
YI
X
integrator
Y
li
XVII
II ~I +U
A/D- or D/A converter
current sensor
volt age sensor
x V III R
Introduction
L
(/)1~
m
[!fi L% ~ dt
u Ri
1
!e
U
= R 2· + L dtdi + e = 'fi!
'fi2
T
Ir? o
'1'1 t( ~ volt age arrowS indicating volt age sources
(u, e) or volt age drops (R i, L ~) 1('pl(~Sent the differences of electrical potential, pointing from the higher to I. h( ~ lower assumed potential. Hence the volt ages in any closed mesh have zero
:: 11111 ,
L: u =
O.
Energy is the basis of any technical and industrial development. As long as only human and animal labour is available, a main prerequisite for social progress and general welfare is lacking. The energy consumption per capita in a country is thus an indicator of its state of technical development, ex hibiting differences of more than two orders of magnitude between highly industrialised and not yet developed countries. ln its primary form, energy is widely distributed (fossil and nuclear fuels, hydro and tidal energy, solar and wind energy, geothermal energy etc.), but it must be developed and made available at the point of consumption in suitable form, for instance chemical, mechanical or thermal, and at an acceptable cost. This creates problems of transporting the energy from the place of origin to the point of demand and of converting it into its final physical form o ln many cases, these problems are best solved with an electrical intermediate stage, Fig. 0.1, where the bold numbers refer to the European grid, because electricity can be
pri7;{J _ . ;ower
------L T;ansmiSSiOntPOWer electronics+consume~ Final I
ene~tat,on T,stobuflon
Fuelcel/s Solar (PV) 46% Fossil--..... 84% 100% "~ 38% Nuclear_ Thermal _ Mechanical _Elektrical Solar --- -~ 'o,Uo=con Biomass' Water, Wind, Waves
/
Electrical drives Control/ad electrical árives
f-
'1' UI = variable ;.>, .
< 16%,« 1% UCTE- European Grid > 350 GW, 1 800 TWh/a
i
Elsctrical ....... Mechênical >50%
Electrical energy per capita and year world wide: 0.02 - 28 MWh
_
Mechanical
Electrical _Thermal _Chemical
Fig. 0.1. From primary energy to final use, a chain of conversion processes
~
Introduction Introduction
• ~enerated from primary energy (chemical energy in fossil fuel, potential hydro energy, nuclear energy) in relatively efficient generating stations, • transported with low los ses over long distances and distributed simply and at acceptable cost, • converted into any final form at the point of destination. 'I'his ftexibility is unmatched by any other form of energy. Of particular importance is the mechanical form of energy which is needed i Il widely varying power ratings wherever physical activities take place, involv ill~ the transportation of goods and people or industrial production processes. I,'or this final conversion at the point of utilisation, electromechanical devices i II the form of electrical drives are well suitedj it is estimated, that about half I.he clectricity generated in an industrial country is eventually converted to Ill( ~ chanical energy. Most electrical motors are used in constant- speed drives I.hat do not need to be controlled except for starting, stopping or protection, 1)11 t there is a smaller portion, where torque and speed must be matched to I.he need of the mechanical loadj this is the topic of this book. Due to the progress of automation and with a view to energy conservation, the need for col1trol is likely to become more important in future. As an example, Fig. 0.2 ~h()ws the mechanical power needed by a centrifugal pump, when the flow is couLrolled by a variable speed drive or, still in frequent use, by throttle and !).\' pass valves.
Pm Po 1,0
l
Bypass ~t_c~,,!~ary_se,,-e~ _ _ _ __
l 'rt<í '-~---, 3' I
( )
Centrifugai pump
Bypass
• Electrical drives can be designed to operate indefinitely in alI four quad rants of the torque-speed-plane without requiring a special reversing gear, Fig. 0.3. During braking, i.e. when operating in quadrants 2 or 4, the drive is normally regenerating, feeding power back to the line. A comparison with combustion engines or turbines makes this feature look particularly attracti ve.
(~ J Motor)
mM
) ) () Load)
_
mL
Throtlle
"I
• Electric drives are adaptable to almost any operating conditions such as forced air ventilation or totally enclosed, submerged in liquids, exposed to explosive or radioactive environments. Since electric motors do not require hazardous fuels and do not emit exhaust fumes, electrical drives have no detrimental effect on their immediate environment. The noise leveI is low compared, for instance, with combustion engines. • Electric drives are operable at a moment's notice and can be fully loaded immediately. There is no need to refuel, nor warm-up the motor. The ser vice requirements are very modest, as compared with other drives. • Electrical motors have low no-Ioad losses and exhibit high efficiencYj they normally have a considerable short-time overload capacity. • Electrical drives are easily controllable. The steady state characteristics can be reshaped almost at will, so that railway traction motors do not require speed-changing gears. High dynamic performance is achieved by electronic controI.
W,
mecham"cal input power
mM mL
Thro",e
= Motor torque
=Load torque
at constant speed V-úJ
Load
variabfe speed
, O
l-..
Flow
1,0
00
Fig. 0.2. Mechanical input power to a centrifugai pump IIKillg difrcrent methods of fiow controi
tM - mM
'J'hc predomil1ance of electrical drives is caused by several aspects: • I':kdric drlv( ~s aT<' available for any power, frOlIl 10° W in electronic wnt.(·II('~i 1.0 I ()H W for drivill ~ plllllpS iII hydro hl.orHI~t' lílnllt.f!, • 'I'1L< ~'y e nV I ' .' tL wid,' rI,W I',C' or I.orqlll' 11.11(1 " p"( 'd, . IO'f NIII , ror il,\I ore IHill lIud,"I' , • 10)' ' /I l1itr , C m l i (·(·l1l.riflll ( · driv(· .
3
Fil';. ().:\. 0l" ~ ral.illg ma d es of nll d('ctric drive: iII 11 11 'f"il,c1 rr,"f.,; (lI' 1.11<' l.o rqllc:j" p c't· c1 plallCi
1
Introduction
Introduction
• The rotational symmetry of electrical machines and (with most motors) the smooth torque results in quiet operation with little vibrations. Since there are no elevated temperatures causing material fatigue, long operating life can be expected. • El ectrical motors are built in a variety of designs to make them compatible with the load; they may be foot- or fiange-mounted, or the motor may have an outer rotor etc. Machine-tools which formerly had a single drive shaft and complicated mechanical internal gearing can now be driven by a multitude of individually controlled motors producing the mechanical power exactly where, when and in what form it is needed. This has removed constraints from machine tool designers . III special cases, such as machine-tools or the propulsion of tracked vehicles, linear electric drives are also available.
Signallevel
I Power levei
-+ I
Voltages, currents
I
~
Torque, acceleration, speed, angular position etc
Signal acquisition:
Contrai: Coordination: Identification: Self tuning: A daptation: Optimisation:
A
Reconstruction of unmeasurable quantities, Elimination of sensors Oecoupling, feed - forward, limiting Coordinate transformation Estimation of variable parameters Computer aided commissioning Adjustment of contrai parameters Minimisation of objective functions
.c .-=:
~
Fig. 0.4. Digital control structure of an electrical drive
1\ s wo u lei Iw IH ' :-; lIppkllwllt, (~d
( ~x p c cted ,
this long list of remarkalJle characteristics is to by Jisa.c1va.ntages of electric drives which lirnit or preclude
U1I·ir 11 SC:
• 'I'1!t'. dqH'lIdHII("( '
011 it COJlLiIlIIOll:-l POW('I'
(' II' !II'"p"I: :i
I...
' ·III' r l , ·.\. ""
UI'
sllpply
C'ill lllPS [)I·.. hl n lll !-J
(,id,I'llilry lli lI,vall n l lk ,
/l ll
witll vdti..
I ·II'r(,rir 1:1U :I'J-( Y
1'''.'\1'<1, whi<-h ln ll il il rdl,Y IlItlllY. I .... " vy
n llel
" XIH'Il
ntv" ( 11 1.0' " 1"'" II,n U.' " y, I
5
or turbine, fuel- or solar cells). The lack of a suitable storage battery has so far prevented the wide-spread use of electric vehicles. The weight of a present day lead-acid battery is about 50 times that of a liquid fuel tank storing equal energy, even when taking the low efficiency of the combustion engine into account. • Due to the magnetic saturation of iron and cooling problems, electric mo tors are likely to have a lower power-to--weight ratio than, for instance, high pressure hydraulic drives that utilise normal instead of tangential forces. This is of importance with servo drives on-board vehicles, e.g. for position ing the control surfaces of aircraft. The electromechanical energy conversion in controlled drives is subject to the terminal quantities of the electrical machine which can be changed with low losses by controllable power electronic converters consisting of semicon ductor switches; they can produce voltages and currents of almost any wave form as prescribed by control which today is executed by microelectronic components. Thus semiconductor technology, combining power conversion and high speed signal processing at reasonable cost, has been the essential force behind the development of todays high performance drives; this is part of the general transition from analogue to digital control systems using mi crocomputers and signalprocessors. Fig. 0.4 gives an impression of how the mechanical, power electronic and control functions, combining hardware and software, are interleaved in a modern drive system.
1. Some Elementary Principies
of Mechanics
Since electrical drives are linking mechanical and electrical engineering, let us recall some basic laws of mechanics.
1.1 Newtons Law A mass M is assumed, moving on a straight horizontal track in the direction of the s-axis, Fig. 1.1 a. Let IM(t) be the driving force of the motor in the direction of the velocity v and h(t) the load force opposing the motion, then Newtons law holds d (Mv) dt
1M - h = -
=M
dv dt
dM
+ vdt' -
(1.1)
where M v is the mechanical momentum. Usually the forces are dependent on velocity v and position s, such as gravitational or frictional forces. li the mass is constant, M = Mo = const., Eq. (1.1) is simplified,
1M - h
dv
(1.2)
= Mo dt ;
with the definition of velo city v = ds I dt, this results in a second order differ ential equation for the displacement,
tM
M
(f(
m,mM
s
J
&
V /:1 I"il-';. J .1 . 'I'n.. ,.. hd.;oll .d Jl llcl "" ULI.;oll a l
b 11101.;011,
"r
IaIlIlP"cI llm..;S(' ~
1. Elementary PrincipIes of Mechanics
~l
fM
!L
-
d2 s = Mo dt 2
1.2 Moment of Inertia
(1.3)
'
"'I .(( \) ..
wh( ~ r e
d2 s dt
dv dt
(1.4)
a = -= 2
is I.he acceleration. If the motion is rotational, which is usualIy the case with ('h:cLrical drives, there are analogous equations, Fig. 1.1 b, d 1I/.M - mL = dt (Jw)
=J
dw dt
dJ
+ w di '
(1.5)
wil.h mM being the driving- and mL the load torque. w = 21l" n is the angular vdocity, in the folIowing called speed. J is the moment of inertia of the 1'( d.:d:iIlg mass about the axis of rotation, J w is the angular momentum. The 1,'1'111 w (dJ/ dt) is of significance with variable inertia drives such as centrifuges 01' n :eling drives, where the geometry of the load depends on speed or time, or illdllst rial robots with changing geometry. ln most cases however, the inertia C; lJI ue assumed to be constant, J = J o = const., hence dw
mM -
mL = Jo di
.
(1.6)
Wit.h E the angle of rotation and w = dE / dt the angular velocity, we have mL = Jo dt 2
(1.7)
'
(ft
i:: Ih( '
d2E dt 2
(1.8)
'
;\.I1i ~ \Ilar
acceleration. II. ~; II()\Ild be noted that mM is the internal or electrical motor torque, not id( 'llli cal with the torque available at the motor shaft. The difference between i111"rllal torque and shaft torque is the torque required for accelerating the il ... rlia of Lhe motor itself and overcoming the internal friction torque of the III< 1101'. 'I'ra nslational and rotational motions are often combined, for example in v(:hich~ propulsion, elevator- or rolling mill-drives. Fig. 1.2 shows a mechanical II1 0dd , where a constant mass M is moved with a rope and pulIey; when 1I<'I',kct.iIlg the mass of the pulley and with '/lI' M \V('
= '/' fM ,
und
mL = r fL
v
= rw
tM
vr
M
- ---=--. (
M
' úJ
-"'r
---. v
Fig. 1.2. Linking linear and rotational motion
1.2 Moment or Inertia The moment of inertia, introduced in the preceding section, may be derived as follows: A rigid body of arbitrary shape, having the mass M, rotates freely about a vertical axis orientated in the direction of gravity, Fig. 1.3. An element of the mass dM is accelerated in tangential direction by the force element dfa, which corresponds to an element dma of the accelerating torque dm = a
l'
d+ =
l'
~a
dv dw dM - = '1'2 dM - . dt dt
The total accelerating torque follows by integration ma
=
wh('n~
(lw
G
/
mL '~r
M
j dma
=j
O
(r
L
t
ma
d2E
'II/.M -
L
/
'1'2
~ dM .
(1.10)
O
Due to the assumed rigidity of the body, alI its mass elements move with the sarne angular velocity; hence M
= dWj 2dM=Jdw ma dt r dt .
(LU)
o The moment of inertia, referred to the axis of rotation, M
J =
j
r
2
dM
(1.12)
o
~ dm . a
I
Jiud wilh M = consto 'II/.M
1/I. J.
'/'
a di,
(M'II) = M
z dw
l' -
di.
'2
(1.9)
.
1\'/ .,.:' 1'1'111'1': :"111.: : til,' ('qllivilklll. 1110111('111. "r illl'l'lin, 01' I.h,' liI)(:arly UIO V illl.~ I l. x i ll (Ir U", plIlI, 'y. "I'P'""III.I V 1.1... 111:1 :::: f\;1 ( 'a,1I II<: 1.1I<1I1J ·,. hl. "I /i') 1,,,1.1( ' ,[ j,iI,! 11 1111., '" " I"" r~ 1.11" , ll'('lIlId"" '1I1 "I"!.II<' \\,11<'1,1 :, (I
p
'1
./,.
rtllI )I::, l',·t'<-n:,'d 1." 1,1...
9
Fig. 1.3. M II IIII·/I1.
"r ill.... l.ia
Fig. 1.4. MCl\wlni of inertia of (·. 'ylilHlcr
cOIlc<·"Ll'i('
1. Elementary PrincipIes of Mechanics
lO
1.3 Effect Df Gearing
i~
a three-dimensional integral. III many cases the rotating body possesses rotational symmetry; as an (')(;tlllple, consider the hollow homogeneous cylinder with mass density º,Fig. I ,ti, As volume increment dV we define a thin concentric cylinder having the r;tdius r and the thickness dr; its mass is
º dV = º 2 1f r l dr .
dM =
M
.l c
f
f
r dM
= º 21f i
3
r dr =
%º i (r~ -
(1.13)
rt) .
G = ºgl1f(r~ -
ri)
(1.14)
l"('slIlts in
r~ +d G 2 9G --2= 9 ri ,
(1.15)
is the gravitational acceleration. The quadratic mean of the radii 1
I ;;
c:t1I, ~ d
7 a
2
dr
+
o
2 r dr]
o
a) 2]
M 12 [1+3 ( 1-2y = 12
(1.17)
.
The minimum inertia is obtained, when the rod is pivoted at the centre, Fig. 1.5 b.
TI
IlIl.roclucing the weight of the cylinder
'ri
~ [J r
r dM =
o
1l"!lce the moment of inertia increases with the 4th power of the outer radius.
wl!('rc.ll
1
2
r2
2
o
J =
Another example is seen in Fig. 1.5 a, where a homogeneous thin rod of length i and mass M is pivoted around a point P, the distance of which from one end of the rod is a. With the mass element dM = (M/i) dr we find for the moment of inertia
1 =
'I'bis redu ces the volume integral to a simple integration along the radius,
11
2 (ri + r~)
(1.16)
1.3 Effect of Gearing Many applications of electrical drives call for relatively slow motion and high torque, for instance in traction or when positioning robots. Since the tangen tial force per rotor surface, i.e. the specific torque of the motor, is limited to some N/c:m 2 by iron saturation and heat los ses in the conductors, the direct coupling of a low speed motor with the load may result in an unnecessarily large motor. It is then often preferable to employ gears, operating the motor at a higher speed and thus increasing its power density. This also affects the inertia of the coupled rotating masses. ln Fig. 1.6 an ideal gear is shown, where two wheels are engaged at the point P without friction, backlash or slip. From Newtons law it follows for the left hand wheel, assumed to be the driving wheel,
Lhe radius of gyration; it defines the radius of a thin concentric
"ylill
J
mM1 -
rI
fI
dWl
= li dt '
(1.18)
where iI is the circumferential contact force exerted by wheel 2. lf there is no load torque applied we have, correspondingly, for wheel 2
r2 fz = lz dw2 dt .
Ml2 3
(1.19)
h is the force driving wheel 2. Since the forces at the point of contact are in balance and the two wheels move synchronously,
Ml2 12
h=h, ( ~ Iimination
O b
[I
1,'1,,;. I.r., M"I","I,
"r \ '"' 1' 1. 1"
"I' (\
,."eI, piv",-,'d
_J~
2 0111. ,,(" , " '111.,-('
of h,
1 ~mMl
rI Wl
=
= r2 W2
(1.20)
,
h, W2 results in
dW l dt
.lI -
tlWI
.II,' dI.
ri
dW2
+ r2 - 1 2 - dt =
[
II
rI
1dWI
2 lz + ( -) 1'2 dt
(1.21 )
1. Elementary PrincipIes of Mechanics
12
1.4 Power and Energy
dw wmM=wmL+Jwdj'
v
13
(1.26)
where PM = w mM is the driving power, PL = w mL the load power and J w (dw / dt) the change of kinetic energy stored in the rotating masses.
~IIJII :
2r3
M3
F ig. 1.6. Effect of gearing on inertia
h + (::) 2 J 2
= J 1 + (::) 2 h + (::) 2
[J3
+ M3 r~l
'
mMl
= J 1e -d
t
W3
+ -Wl
r3g M 3
.
(1.24)
The rotational motion of the mechanical arrangement shown in Fig. 1.8 is dC!icrihcd by a first. order dilferential equation for speed Ii.w
w. /, I .I ti,
Mlll !. q,l ill d ,IIIII I,y
( 01
,vi.. ldn UI<' PClW(" 1.,11 " 11('('
)-)
mL ...... ....
PL
Fig. 1.8. Power flow of drive
By integrating Eq. (1.26) with the initial condition w (t the energy input t
WM(t) =
f f
t
o
o
= O we find
t
PLdT
+
o
f
Jw: dT
o
w
t
=
f + f
PMdT =
= O)
PL dT
J
n dn
o
(1.27)
(1.23)
1.4 Power and Energy
'III'M
J
1 2 =wL(t)+2"Jw.
illduding the equivalent inertia of the mass M3 being moved in vertical direc lion. Applying Newtons law, taking the load of the hoist into account, results iu dWl
5EJ
(1.22)
,
which indicates that a rotating part, moving at higher speed, contributes 1I1Ore strongly to the total moment of inertia. ln Fig. 1.7 an example of a multiple gear for a hoist drive is seen. J l , .J2 , Js are the moments of inertia of the different shafts. The total effective illertia referred to shaft 1 is
J 1e
'W
PM -
Fig. 1. 7. Hoist drive with gear
J le is the moment of inertia effective at the axis of wheel 1; it contains a component reflected from wheel 2. ln most cases it is easier to determine the speed ratio rather than the radii, J le =
M
m
(1.25)
The last term represents the stored kinetic energy; it is analogous to 1 1 1 2 -Mv 2 - Li 2 -Cu 2 . 2 ' 2 of other energy storage devices. Since the energy content of a physical body cannot be changed instantaneously - this would require infinite power the linear 01' rotational velo city of a body possessing mass must always be a continuous function of time. This is an important condition of continuity which will frequently be used. Because of the definitions ds de v= and w = dt dt the positional quantities s and € are also continuous functions of time, due to fiuite sp eed. This is also understandable from an energy point of view, since position lIlay be associ ated with potential energy, as seen in Fig. 1.7, where I.h, : Ill as ~; 1I{, i:-l ]>os il.iollf:(} vcrtically dC(l':lIdillf!: OH th,: angle of rotation of 1.11<' d ri Vl' sJlid'i. .
1. Elementary PrincipIes of Mechanics
14
1.5 Experimental Determination of Inertia
1.5 Experimental Determination of Inertia
ú)
The moment of inertia of a complex inhomogeneous body, such as the rotor of an electrical machine, containing iron, copper and insulating material with complicated shapes can in practice only be determined by approximation. The problem is even more difficult with mechanicalloads, the constructional details of which are normally unknown to the user. Sometimes the moment of inertia is not constant but changes periodically about a mean value, as in the case of a piston compressor with crankshaft and connecting rods. Therefore experimental tests are preferable; a very sim pIe one, called the run-out or coasting test, is described in the following. Its main advantage is that it can be conducted with the complete drive in place and operable, requiring no knowledge about details of the planto The accuracy obtainable is adequate for most applications. First the input power PM (w) of the drive under steady state conditions is measured at different angular velocities w and with the load, not contributing to the inertia, being disconnected. From Eq. (1.26),
PM
= PL + J w
dw dt '
(1.28)
the last term is omitted due to constant speed, so that the input power PM corresponds to the los ses including the remaining load, PM = PL. This power is corrected by subtracting loss components which are only present during power input, such as armature copper los ses in the motor. From this corrected power loss P~ the steady-state effective load torque m~ = p~jw is computed for different speedsj with graphical interpolation, this yields a curve m~ (w) as shown in Fig. 1.9. For the run-out test, the drive is now accelerated to some initial speed wo, where the drive power is switched off, so that the plant is decelerated by the loss torque with the speed measured as a function of time, w(t). Solving the equation of motion (1.25) for J results in '" -m~(w) J '" dw '
mM=O.
(1.29)
âf(w)
Rence the inertia can be determined from the slope of the coasting curve, as shown in Fig. 1.9. Graphical constructions, particularly when a differentiation is involved, are only of moderate accuracy. Therefore the inertia should be computed at different speeds in order to form an average. The accuracy requirements regarding inertia are modest; when designing a drive control system, an error of ± 10% iil IlsualJy Itcccptable without al1y ~criom; effect. Two spcc.j;d CiLS(': ; }PaI! to [>itrticlllarly sill1pl(~ iJlt.(:rprüt,ations: ii,) Ã n:i11 III i "I'.
1.11 (1 1" 01"1,,,(' 1,'.1 IOI'!Il t.OI·qlle' '111,'1. to !lo Il,pproxilJlat.dy (:OIlK(:i\.Ut.
Jilldl. \,d Iii>!"'" ,"I,,·, vld ,
15
ill n.
coasting curve
oo(t)
Corrected steady state load curve m'L (00)
m'
L
00,
, m'L (00 )
1
Fig. 1.9. Run - out test m~ ~ const,
for
Wj
< w < w2
,
then w(t) resembles a straight line; the inertia is determined from the slope of this line. b) li a section of the loss torque may be approximated by a straight line, m~ ~ a+bw,
for
Wj
< w < W2
.
a linear differential equation results,
dw
J dt +bw= -a.
The solution is, with
w(t) = -~
b
+(w
2
W(t2)
=
W2,
+~) e- b (t-t2)/J b
'
t
2:
t2 •
Plotting this curve on semi-Iogarithmic paper yields a straight line with the slope -bj J, from which an approximation of J is obtained.
2. Dynamics of a Mechanical Drive
2.1 Equations Describing the Motion of a Drive with Lumped Inertia The equations derived in Chapo 1
dw J dt = mM (w, dE: dt
é,
(2.1)
YM,t) - mL (w, é, YL,t) ,
(2.2)
= w,
describe the dynamic behaviour of a mechanical drive with constant inertia in steady state condition and during transients. Stiff coupling between the different parts of the drive is assumed so that all partial masses may be lumped into one common inertia. The equations are written as state equations for the continuous state variables w, é involving energy storage, e. g. [38,88]). Only mechanical transients are consideredj a more detailed description would have to take into account the electrical transients defined by additional state variables and differential equations. The sarne is true for the load torque mL which depends on dynamic effects in the load, such as a machine tool or an devator. AIso the control inputs YM, YL to the actuators on the motor and load side have to be included. Fig. 2.1 shows a block diagram, representing the iuteractions of the mechanical system in graphical formo The output variables of the two integrators are the continuous state variables, characterising the (~l1ergy state of the system at any instant. Linear transfer elements, such ;\.S integrators with fixed time constants, are depicted by blocks with single !"rames containing a figure of the step response. A block with double frame ,lCllominates a nonlinear function; if it represents an instantaneous, i. e. static, Ilonlinearity, its characteristic function is indicated. The nonlinear blocks i Il Fig. 2.1 may contain additional dynamic states described by differential ('qllations. Dcpenc\cIlce of the driving torque mM on the angle of rotation is a char ad(~ristic fCiI.tUfe of synchronous motors. However, the important quantity is II' d. Lhl' allgl!' of roCaLiol\ é itself but the difrc r('uc( ~ itllgle 8 against the no-load 1(( w ii ich is dd(~rIlJiu(~d hy 1.111' l.o('(pl(· . () ud('r 1.1«' loa.! 11111',1(, S iH (,oll:d.alll..
11.111,:1"
~lt.('ady statl'
condi I.io/l s
2.1 Equations Describing the Motion of a Drive with Lumped Inertia
2. Dynamics of a Mechanical Drive
li>
However, when the partial masses are coupled by flexible linkages, such as with mine hoists, where drive and cage are connected by the long winding rope or in the case of paper mill drives with many gears, drive shafts and large rotating masses particularly in the drying section, a more detailed description becomes necessary. The free oscillations may then have frequencies of a few Hertz which are well within the range of a fast controlloop. ln Fig. 2.2 an example is sketched, where the drive motor and the load having the moments of inertia J 1 , J 2 are coupled by a flexible shaft with the torsional stiffness K. The ends of the shaft, the mass of which is ignored, have the angles of rotation C1, C2 and the angular velocities W1, W2 . Assuming a linear torsionallaw for the coupling torque me,
Power
supp/y
YM
o
Motor
co
mM
mL
me Fig. 2.1. Simplified block diagram of lumped inertia drive
me
~1fl>, Motor
))K)(1 me
Cü2,E2
J2 Load
=K
(2.3)
(C1 - c2) ,
and neglecting internal friction effects, the following state equations result
ln order to gain a better insight, let us first assume that the electrical transients within the motor and the internalload transients decay consider ably faster than the mechanical transients of w and ci as a consequence, it r()l1ows that the motor and load torques mM, mL are algebraic, i.e. instan (,allcouS functions of w, c and 8. Hence, by neglecting the dynarnics of motor ;Lmlload, we arrive at a second order system that is completely described by lhe two state equations (2.1), (2.2). So [ar we have assumed that alI moving parts of the drive can be combined lo forrn a single effective inertia. However, for a more detailed analysis of dy n
/TIM , <Út,f,1
19
)))
mL
Jo'ilf . 2,2. I )riv!' l'1I1I:.i/<(, I.'I: 01' mot,or nlld 101\.1 '-,0111'1".1 1,.1 ""xij,k J;ltart.
dw 1 J 1 dt = mM - me = mM (Wt, C1, YM) - K (c1 - c2) ,
J2
dw
dt2 = me de 1
dt = W1, de 2
-=W2·
dt
mL
=K
(cl - c2) - mL
(W2,
c2, YL) ,
(2.4) (2.5) (2.6)
(2.7)
A graphical representation is seen in Fig. 2.3 a. Here too, the torques mL are in reality defined by additional differential equations and state vdriables. If only the speeds are of interest, the block diagram in Fig. 2.3 b "':1..)' be useful, which, containing three integrators, is described by a third lIJ'dc[' differential equation. With increasing stiffness of the shaft, the quantities C1, C2 and W1, W2 I"'collle tighter coupledi in the limit the case of lumped inertia emerges, w lU'r<: C1 = C2, W2 = W1. Ir Lhe transmission is affected by mechanical backlash, where the two "",rLias separate when the shaft torque changes sign, the linear torsional III " Il('h f{ is replaced by a nonlinear function, as shown in Fig. 2.3 c. This is III' i IlIportance with reversible drives, for instance servo drives. (lltviollSly the subdivision of the inertia may be continued indefinitely; o' Vo-J'y Lime a new partial inertia is separated, two more state variables have to 100 ' ,I, -Ii 11(:(1, t.ransforming Fig. 2.3 into a chainlike structure. A typical example, w l",I'" 1I1,tlly partial masses must be taken into account for calculating stress 111141 I'n.t.igw!, is a turbine rotor. III IV!,
20
2.2 Two Axes Drive in Polar Coordinates
2. Dynamics of a Mechanical Drive
where w = dE; / dt and v = dr / dt are the rotational and radial veloci ties . After separating the terms of acceleration in tangential and radial direction and superimposing frictional and gravitational components, Newtons law is ap plied in both directions, resulting in the equations for the mechanical motion of the centre point of M 2
Power supply
Power supp/y
21
fO
J r,,-_~A
(ll + h
,
dw
+M2r2) dt = Coriolis
Gravitation
,...-"'-...
,.-"'-----....
mM - 2M2 rwv-M2 gr cose:-mF -mL,
(2 .10)
de:
dt = w ,
b
a
dv M 2 dt
c~JP
dr dt = v .
F ig. 2.3. Block diagram of twin-inertia drive with flexible shaft (a) Model of mechanical plant, (b) Reduced order model,
c) Mechanical transmission with backlash
2.2 Two Axes Drive in Polar Coordinates On machine tools or robots there are normally several axes of motion, that rnust be independently driven or positioned. An example is seen in Fig. 2.4 a, where an arm, carrying a tool or workpiece , is rotated by an angle e:(t) around a horizontal axis. The radial distance r(t) from the axis to the cen ter of the mass M 2 represents a second degree of freedom, so that M2 can bc positioned in polar coordinates in a plane perpendicular to the axis. The rotary and radial motions are assumed to be driven by servo motors, produc ing a controlled driving torque mM and a driving force 1M through a rotary gear and a rotary to translational mechanical converter, for instance a lead screw. With fast current control the motors are generating nearly instanta \ICOIlS impressed torques, serving as control inputs to the mechanical planto For simplicity, the masses are assumed to be concentrated in th.e joints, re sltlt.ing in the inertias h, J 2 . The coupling terms of the motion can be derived by (~xpressillg t,he acceh~ration of the mass M 2 in complex formo dr
!l
dI
(li
,(J. )'
,11 ',\
(1' (..1. ' )
,II
. (1< i' ) ~
(2.8)
('II I :i r w) (.3 ~ ,
,, ~\
,
(/L'II d'
I_,.}
.~ ) ;I , .
. (
t' I (,
.\1 "
= 1M +
, '~~ (LI
'fi
I
'I'
IlldIU )
(' j ' '
I
(/..!J )
(2.11) C entrifugai
Gravitation
,-"--.. 2 2r w
,...-"'-...
M
-
M 2 g sine: - lF
h,
(2.12) (2.13)
The equations (2.10)-(2.13) are depicted in Fig. 2.4 b in the graphical form of a block diagram, containing four integrators for the state variables. Despite the simple mechanics, there are complicated interactions, which become more prominent with increasing rotary and radial velocities. The control of this mechanical structure is dealt with in a later chapter. Clearly, the two motions are nonlinearly coupled though gravitational, Coriolis- and centrifugaI effects; they are described by four nonlinear state equations. mF, lF and mL, h are due to friction and external load forces with may exhibit their own complicated geometric or dynamic dependencies, If it is important for the application to express the position of mass M 2 in cartesian coordinates, this is achieved by a polar-cartesian conversion x (t) = r cos e: ,
y(t)
=r
sine: .
(2.14)
(2.15)
Moving the arm also in the direction of the axis of rotation, so that the mass M 2 can be positioned in cylindrical coordinates, would introduce a third
dccoupled degree of freedom . The dynamic interactions for a general motion, involving six degrees of r)'(~edom (three for the position, three for the orientation of the tool) are ex ('cedilJgly complicated, they must be dealt with when controlling the motions 01' Jllult.i ax( ~s rouots with high dynamic performance [M53, 013, 849].
2.3 Steady State Characteristics of Motors and Loads
2. Dynamics of a Mechanical Drive
'22
variable is the load angle 0, i. e, the displacement of the shaft from its no load angular position. When the maximum torque is exceeded, the motor falls out of step; asynchronous operation of larger motors is not allowed for extended periods of time because of the high currents and pulsating torque, The electrical transients usually cannot be neglected with synchronous motor drives. The rigid speed of synchronous motors when supplied by a constant fre quency source makes them suitable for only a few applications, for exarnple large slow-speed drives for reciprocating compressors or synchronous gener ators operating as motors in pumped storage hydro power stationsj at the low end of the power scale are electromechanical clocks, The situation is dif ferent, when the synchronous motor is fed from a variable frequency inverter because then the speed of the drive can be varied freely (Chap. 14). With the progress of power electronics, these drives are now more widely used.
Friction
mLosd
r2
,
M2 V J +J2+ grcos E
M
23
gsinE
I rw 2
y/, r
fLoad
Friction
x
a
b Fig. 2.4. Two axes drive in polar coordinates (a) Mechanical plant (b) Block diagram
2.3 Steady State Characteristics of
Different Types of Motors and Loads
Consider first the steady state condition, when the torque and speed of a single axis lumped inertia drive are constant and the angle changes linearly with timej this condition is reached when mM - mL = O. With some motors, such as single phase induction motors, or loads, for example piston compres sors or punches, the torque is a periodic function of the angle of rotationj in this case, the steady state condition is reached, when the mean values of both torques are equal, mM - mL = O. The speed then contains periodic oscillations, which must be kept within limits by a sufficiently large inertia. The stcady state characteristics of a motor or 1030<1 are often functions given in graphical [o fi II , <:onnN:ting main varíilhl(~K, sllch as speed and torque; lh(' provisíou i:i 1.11;t\. :lllxili
li(.ld
(·II!"l"('IIt.,
Ht.iI.lIl..
I" 1' 11\ '). ,1, /1 ,1,11,1"" 1.,Ypi(,iI.I
mM Fig. 2.5. Steady state torque-speed characteristics of (li) electrical and (b) mechanical drives
The "asynchronous" or "shunt-typ" characteristic in Fig, 2.5 a is slightly clrooping; often there is also a pronounced maximum torque. The lower por Lioll of the curve is forbidden in steady state in view of the high losses. With I.hree- phase asynchronous motors the rotor angle has no effect on the torque i Il steady state. Motors with "series-type" characteristic show considerably larger speed drop under loadj with DC or AC commutator motors, this is achieved by ;1. suitable connection of the field winding. The main area of applications at larger ratings are traction drives because the curved characteristic resembling ;1. hyperbola facilitates load sharing on multi pie drives and permits nearly collst.ant power operation over a wide speed range without gear changej this is par\.icularly suited to a Diesel-electric or turbo- electric drive, where the 1',,11 !)()Wl'r o[ lhe thermal engine must be used. For ('olllparisoll, SOIlJC t:ypical charactcrist.ics of a Lurbine anel a Diesel ('I\/';i ll(' ai. ('ClIlHl.n,lll. flld illjc('.t.ioJl pc r Ht.I'III(' lU'" i'iC'('1\ iII l"il':. :~.S h.
25
2.3 Steady State Characteristics of Motors and Loads 2. Dynamics of a Mechanical Drive
21.
The curves in Fig. 2.5 a are "n~tural" characteristics which can be modi [ied at will by different control inputs, e. g. through the power supply. With dosed loop control a shunt motor could assume the behaviour of asynchronous O[ of a series type motor. As an example, typical steady state curves of a con l.rolled DC drive are shown in Fig. 2.6; they consist of a constant speed branch (normal operation) which is joined at both ends by constant torque sections aetivated under overload condition through current limito Figure 2.7 depicts Lhe steady state curves of the motor for driving a coiler. li the electrical power rcference is determined by the feed velo city v of the web or strip to be wound and includes the friction torque, the coiler operates with constant web force f independent of the radius r of the coil, PL = vi + PF· The steady state characteristics of mechanicalloads are of great variety; however, they are often composed of simple elements. This is seen in Fig. L.8 with the example of a hoist and a vehicle drive. The gravitational lift Lorque mL is independent of speed (Fig. 2.8 c); in the first quadrant the load is lifted, increasing its potential energy, hence the drive must operate in the Illotoring region. ln the fourth quadrant, the power flow is reversed with the load releasing some of its potential energy. Part of that power is flowing back Lo the line, the remainder is converted to heat losses. The lower half of the winding rope, seen in Fig. 2.8 a, serves to balance the torque caused by the
ú)
m = const
1 L __ _
ú)tl
r"
= r:onst m =const
Ú)2
Torque limit
mL = r 9 (MI
- M2 )
(J)
Gravitational torque
mL mL = r fL
= r9 M sin a
a
c
b
Fig. 2.8. Drives involving gravitational forces
weight of the rope; this effect can be substantial on a winder for a deep mine, tending to destabilise the drive. All mechanical motion is accompanied by frictional forces between the surfaces where relative motion exists. There are several types of friction, some of which are described in Fig. 2.9. ln bearings, gears, couplings and brakes one observes dry or Coulomb friction (a), which is nearly independent of speed; however one has to distinguish between sliding and sticking friction, the difference ofwhich may be considerable, depending at the roughness ofthe surfaces. The forces when cutting or milling material also contain Coulomb type friction. ln welllubricated bearings there is a component offrictional torque which rises proportionally with speed; it is due to laminar flow of the lubricant and is called viscous friction (b). At very low speed and without pressurised lubrication, Coulomb type friction again appears.
m Fi~. 2.6. Torque/speed curves of controlled motor
wiLlt constant speed branch and torque limit
a
ú)
/b ~
/./
ú)
/
i/ /
~
---,
/// /// /..,./
__ --:/' r/'lM
[)
lt\i,-,. ~ . 7 . ('~Iil~ll' dll\'(~ ('1) 1\~1·(" IL·H I(('~ 1 lIllel (II)
I d, f,t,k
IllIHI
/
Fi~. ('IIl"V(':1
a
~
mG
/
mL
mL
I
II Gravltatlona .. I
r----
torque
I
/
n
ú)
/ ......,c
2.U. llirrnnnL typcs of frictional Fig. 2.10. Torque/speed curves of a hniHI. d ri v('
l.o!''I''I:
26
2. Dynamics of a Mechanical Drive
2.4 Stable and Unstable Operating Points
With pumps and ventilators, where turbulent flow occurs, the torque rises with the square of speed; the drag force due to air flow around vehicles or l,!Je torque required by a centrifugaI blower pushing cooling air through an clectric motor also have this characteristic (c). ln practical drives, with the motor as well as load, all these types of friction cxist simultaneously, with one or the other component dominating. When driving a paper mill, printing presses or machine tools, Coulomb friction is usually the main constituent but with centrifugaI pumps and compressors the lorque following the square law is most important, representing the useful lllechanical output. Note that frictional torques are always opposed to the direction of relative motion. ln Fig. 2.10 various torques, acting on a crane under load, are drawn; me is the constant gravitational torque caused by the load of the crane, mF is the Cri ctional torque, resulting in the totalload torque curve (a). The speed Wo corresponds to the run- away speed with no external braking torque. When for safety reasons a self-Iocking transmission, such as a worm gear (b), is eruployed the crane must be powered even when lowering the load; this is du e to the large sticking friction.
or in normalised form,
J dLlw
k
dt
+ Llw
= O
where
k= -
â
(mL - mM)
âw
"t ,"", ", .": =='1(
00
O
--;/'
00,
mL
mL
w,
= mM (w, t) - mL (w, t) .
/
m
(2.16)
Apparently, a steady state condition with a constant rotational speed possible, if the characteristics are intersecting at that point, i. e. for mM
(Wl) - mL (WI) =
Wl
I
,1 (cl ,
.I
wc
d !\ (,) ,/I
littd 1.1\(:
o.
lI'111 nl
IIt .J
lill('aris(~d
l ,\,,) 'u
I
('quat.iotl
Ihll.J, }(4 )
I 1. 11
!\/,; ,
m
mM
L
m
k=O
stable
k
indifferent
a
m
unstable
b
c
Fig. 2.11. Stable and unstable operating points
cu
L3
cu
3/ ___...... m --J.
unstable
M
2/
~
~
L2
mL
~
mM stable
~
is
III order to test wltether this condition is stable, Eq. (2.16) can be linearised ai I.h(: o])(:ratitt /-Ç poiltt W I, assumiug a smn.ll displa(:ernent Llw. With G:J = (ti l
/
/
)
di
- - -/
W,
~
k> O
J
(2.17)
+4:t __ ~~4W<:/
mM
2 .4 Stable and Unstable Operating Points
dw
IWl
This is illustrated in Fig. 2.11 with some examples. The assumed steady state is stable for k > O; in this case a small displacement Llw, that may have been caused by a temporary disturbance, is decaying along an exponential function with the time constant T m = J/ k. k can be interpreted as the slope of the retarding torque at the operating point, as seen in Fig. 2.11. For k < O the operating point at Wl is unstable, i. e. an assumed deviation of speed incre ases with time; a new stable operating point may 01' may not be attained. The case k = O corresponds to indifferent stability; there is no definite operating point, with the speed fiuctuating due to random torque variations.
......mM
I~'y ignoring the dependence of driving and load torques on the angle of rota t.iOll, the corresponding interaction seen in Fig. 2.1 vanishes and so does the df'('ct of Eq. (2.2) on the static and transient behaviour of the drive. Equa I.joll (2 .2) then becomes an indefinite integral having no effect on the drive. I r wc also neglect the electrical transients and the dynamics of the load, the n:lllaillillg mechanical system is described by a first order, usually nonlinear, d iífcrential equation but, in view of the simplifications introduced, its validity is rcstricted to relatively slow changes of speed, when the internal transients alld interactions in the electrical machine and the load can be neglected.
27
m
m
Fig. 2.12. Induction motor with differ- Fig. 2.13. Rising torquejspeed curve as ent types ar load cause of instability
Figure 2.12 depicts the steady state characteristic of an induction motor (7I/.M) /.og d.l\(:r with some load curves (Sect. 10.2). LI could be the character i ~ l.i(' Or ; l vClll.il a tiug fali ; thc intersedioll ] is st.ahlf:, rOltghly corrcsponding to IICIIll i ll HI I"ad . Wil.h IJ~ 1.I11~J'(' is N.l S O ;t fi l.nl)I, · "1'<')'ni..iJlI '; pnill(.) bltt. tlw 11101.0)'
11'1
2. Dynamics of a Mechanical Drive
would be heavily overloaded. With the ideallift characteristic L3, there is an IIllstable (3) and a stable (3') operating point, where the motor would also h(' overloaded. In addition, the drive would fail to start since the load would pllll the motor in the lowering direction when the brakes are released and I.!lere is insufficient frictional torque. A particularly criticaI case is seen in Fig. 2.13. A slightly rising motor characteristic, which on a De motor could be caused by armature reaction dll( ~ to incorrect brush setting, leads to instability with most load curves n ccpt those intersecting in the shaded sector. The stability test based on a linearised differential equation does not fully (:xdude instability if electrical transients or angle-dependent torques should II:tvc been included. The condition k > is only to be understood as a Iwccssary condition, even though it is often a sufficient condition as well.
°
3. Integration of the Simplified Equation of Motion
With the assumptions introduced in the preceding section the motion of a single axis lumped inertia drive is described by a first order differential equation, Fig. 3.1,
J
dw
di = mM (w,
= ma (w,
t) - mL (w, t)
(3.1)
t) ,
which upon integration yields the mechanical transients. Several options are available for performing the integration.
mM'
úJ, ê
=E) ~
~- ~} mL
Fig. 3.1. Drive with concentrated inertia
3.1 Solution of the Linearised Equation In the linearised homogeneous equation, Eq.(2.17), d(Llw) Tm~
+ Llw
= 0,
Tm
J
is Llw a small deviation from the steady state speed
k = ii.
(3.2)
=k Wl
and
~ (mL - mM)1 âw
(3.3) w]
II 1<:
.I / J..:
1111 1'
UI("
IIW:tllilll', of a IlIcchalliraJ tilli(' rolii il.nlll..
'I'll!"
point. T m
=:
1~( ~ II( ~ ral KO!lll.ioll iK
3. Integration of the Simplified Equation of Motion
:30
L1w(t) = L1w(O) e- t / Tm
3.1 Solution of the Linearised Equation
The slope of the retarding torque is
(3.4)
where L1w(O) is the initial deviation, that may be caused by switching from one motor characteristic to another resulting in a new steady state speed. Because of the stored kinetic energy, the speed is a continuous state variable (the only one because of the simplifications introduced).
k
= -mo > O; Wo
(mL - mM)
+ L1w = O,
d(L1w) Tm~
mL
(3,5)
where
Jwo
Tm
---,mM1
;\w(O)
â = -âw
hence the differential equation is stable, it has the form
,ócu
cu
31
mo
is the mechanical time constant.The steady state operating point at the in tersection of the two torque curves is Wl = wo; due to the initial condition at standstill, we find L1w(O) = -Wo, which leads to the particular solution
a
m
b
Tm
F ig. 3.2. Mechanical transient of linearised drive system
L1w(t)
or wi th w = Wo
w(t) Til Fig. 3.2 a it is assumed that the motor, initially in a steady state ('(Jlldition at point 1, is switched to another supply voltage so that a new ('iJaraderistic for the driving torque mM2 is valido This causes an initial devi ;i.I.io ll L1w(O) with respect to the new operating point. The deviation vanishes IIlolI/,. ; LII exponential function (Fig. 3.2 b), the time constant of which is de1."l'llIill pd hy the inertia and the slope of the load- and the driving-torques. 'l'hi N is di scu::ised with the help of some simple examples.
:1.1. 1 St.art of a Motor with Shunt-type Characteristic at No-Ioad
t = O, Fig. 3.3. The of the motor is assumed to be linear between no-load liI)('( 'd WI) (lIeglecting friction load) and stalled torque mo'. With normal mo 1.111:: or lIwdium size, the stalled torque may be 8 or 10 times nominal torque; i I. is dd.cnnined by extrapolating the drooping shunt-type characteristic to Hla lld sLill anel hence represents only a reference quantity that cannot be mea :: lIl'cd iII pmctice. However, in the present example we assume that a starting 1'(·::i::(.OI" lia.s been inserted, reducing the stalled torque to perhaps twice nomi lI a l Lorqlw alld thus extending the validity of the linear characteristic down to HL;\.lIdsLill. This, by the way, m akes the simplifying assumptions more realistic :l llIn : Lhe dectric;J.l tmn sicllts becornc faster while the m echanical transients fi. IJlolor which initially is at rest is started at
= -Wo e- t / T= + L1w
= Wo (1 -
e- t / T =) .
(3.6)
From the torquejspeed curve of the motor ~
Wo
= 1- mM
(3.7)
mo
the motor torque during the start- up is obtained (Fig. 3.4), mM(t)
= mo e- t / Tm
.
(3.8)
The discontinuity of the motor torque is caused by the omission of the elec trical transients. ln reality the torque is also associated with energy states and, hence, is continuous.
1 .() I'Ill\"jsp( ~c d- curve
cuo'" t~
00
Uort0
1~k--------
mM
- mL=O
=O mà m t
alT
:l lIl ldl v. dll'·:
cu mM cuo' mo
cu
I"il;. :1 .:1. SI.fIl' l.illl'.
"r
1l1ll1.0!'
ai.
110
10ft"
Tm Fil-'( . :1.'1 . St.art,iIl14 t.rall fi it~ lIl,
3. Integration of the Simplified Equation of Motion
:1:2
3.1 Solution of the Linearised Equation
:1.1.2 Starting the Motor with a Load Torque Proportional to Sp eed
w
~ WI'
mM
Wo
mo
1
1
111 Fig. 3.5 the curves of a drive with speed-proportional load torque are displayed. The steady state operating point is now at WI < Wo, hence L1w(O) =
33
m1
W1
Wo
mo
"
úJ
"
,
' . . . . .......m L = O
....
....
{Oo
mM
Tm1
a
Tm
t
Tm1
b
Tm
t
Fig. 3.6. Starting transient with load
(1)1
3.1.3 Loading Transient or the Motor Initially Running at N o-load Speed
m
m1
The motor is now supposed to be initially in a no-Ioad condition at the speed Wo with the starting resistor short circuited, so that the speed droop is reduced and the extrapolated stalled torque assumes its large nominal value mOn· The torque/speed-curve is then linear only in a narrow speed range around no-Ioad speed.
I"ig. 3.5. Starting of motor with linearly rising load torque
The slope of the retarding torque is
kl = -
ô
ÔW
mo
(mL - mM) = -
> O,
(3.9)
WI
úJ
t= O
wlüch leads to a reduced mechanical time constant, JWI
Tml = mo
=
WI
Wo
T
m
.
(3.10)
TIl\' speed transient is
W(t) =
WI
mL
(1 - e- t / Tm ,).
........................................
(3.11)
. . . . . . . . mM ................
1';lilIlination of speed from the motor characteristic Eq. (3.7) results in the. d ri vi11g torque mM(t)
= mo
(1- :J = [1- :~
= mI + (mo Ilol.h
tnl.nsi(~J1l.s af( ~
mo
mI) e- t / T ""
.
t (1- e- / Tm ,)]
........
mOn m
Fig. 3.7. Applying constant load torque to motor running at no-load (3.12)
plol.ted in Fig. 3.6, together with the no-Ioad starting 1.1';t.J1Hi(~J1(. (111./, - O) . AI. I. 0, t.he Rlopc of tltc specd curves is the sal1lC, 1)('("a.lI :ic- tlI(" ;,1<'("(,1('1 n l.illl', l.orqll(,i' ;1.1"(' idcul.ieal n.1. (v · O.
=
m1
At t = O a constant torque load without inertia is applied to the motor iloft, possibly by a mechanical brake, Fig. 3.7, causing the motor to slow dowlI [.0 Lhe steady state speed wI; hence the initial deviation is L1w(O)
Iii I
(')0
W I .
III vil'W
("ow;L;mt
load torqll(',
WI:
lilld
3.1 Solution of the Linearised Equation
3. Integration of the Simplified Equation of Motion
:11
~ (mL âw
k =
mM)
mO n
I
Tm
Wo
Wl
JwO = T mn
= mO n
·
(3.13)
Sillce mO n corresponds to the extrapolated stalled torque with short circuited sl.arting resistor, T mn is also called short circuit mechanical time constant. The solution of Eq. (3.5) is
Llw(t) = (wo - wd e t / Tmn 01'
w(t) =
Wl
+ (wo - wd e- t / Tmn
(3.14)
•
The driving torque is again obtained from Eq. (3.7), mM(t)
= mO n =
mI
(1-
~J
= mO n (1-
(1 - e- t / Tmn )
~:)
(1-
t Tmn e- / )
(3.15)
•
III Fig. 3.8 the transients are drawn for loading and unloading the motor.
mM ..ill..
- wo m- On 1
----1(O
___
..ill..
o
W 1~~-~
(1)0
mi
computer, can be made immune against short time disturbances on the line side, when it is supplied by a rotating motor-generator set having a fl.ywheel attached.
3.1.4 Starting or a De Motor by Sequentially Short-circuiting Starting Resistors With DC motors fed from a constant voltage bus the starting circuit shown in Fig. 3.9 may be used, where the contactors C 1 , ••• CN are closed in succession. By suitable choice of the resistors, the droop of the resulting torquejspeed characteristic is progressively reduced. ln order to achieve fast starting with out overloading the motor, the torque is specified to vary between the limits mI, m2 during the start, as is shown in Fig. 3.10. The armature current, being proportional to torque, then varies between the limits i I, i 2 . A similar starting procedure exists for AC induction motors having a wound rotor and starting resistors in the rotor circuito As the motor is accelerated, the torque mM decreases until the lower limit mI is reached; then the next section of the resistor is short-circuited causing the torque to jump again to the upper limit m2. This is repeated until alI the starting resistors are switched out and the motor operates on its natural torquejspeed curve. ln steady state the torque is determined by the load. The required number of sections of the starting resistor and their appropriate values can be calculated in closed form [37] .
1
IL-
C1
1
mM
m() n
1 1_
=-=-
mOn
C2
Cn
1
1 1_ _ _ _ _ _
Co
I' 1
Tmn
35
t
vai
Re
Le
1"11';. ~L8. Loading transient
'l'l\( ~ drivillg torque apparently follows the applied load torque with a lag
Fig. 3.9. Starting circuit for a De motor
d!'l.(~rlllillcd
by T mn . The reason for this is that during the deceleration phase h Ol1W of til(' kinctic energy stored in the inertia is released and makes up for pMI. 0[' I.he loat! torque; when the drive is accelerated after disconnecting the !O, H!, I.he 1l1ÜiSill~ kinetic encrgy is restored. Thc mcchanical energy storage t.ltll S act.s HS a hl\fr(~r !>dweell lhe load a.nel lhe ek('(,ri(:al sllJ>ply fccding the Tltin dr,'d Ill ay b(: n.cn~Ilt.\1at.(:d hy iI.dtlilll~ It f1yw'llt'd lo Lhe: drive:, (II IIl'Ikl' l.tI 1'1'1",,'(' " tllt' :lllpp! y fiy Ht.(1II1 r""11I h ip,h !, 1I11 1 11111'1( " ;1 Hllrll n~i ti IO/i(> ('1I 1l 1.~ , d 1,1' 111>1'1'." 10 11 1111 ', IlIill d dvl 'rlj """VI1l'lll'l y, II /11' 11 /111. 11', , 1,,1101 , 1'''1 "',11.1111'' ' ii
I HO l.or .
From Fig. 3.10 the following relation is found rn,2 1/1,2
mI
w
(v + 1) - w (v) Wo - W(v)
Wl,idl 1" ':-:1111.::, wil.lt Ul(: ;-\.IJI,rnvia(.joll '11/.1
(3.16)
jm./,
U·
-I, iII a. recllrsiv(' forlllula
:~ (;
3.1 Solution of the Linearised Equation
3. Integration of the Simplified Equation of Motion
37
mM
w (l)(N)
m1
(()(v+ 1)
(ú(v)
(ú(1)
..... ..... ..... ,>,
I
w(O)=O jo'i~. 3.10.
m
m:z=mo{O)
1
•
mdO)
m
: - Tm(I)--: I I I I
Torque/speed-curves for different values of armature resistor
cu W
CUo
(1/ + 1) = (1 - a) Wo + aw (v) = w(l) + aw (v) .
i\SSlIlllÍug w(O)
+ a) w(l) (1 + a + a2 ) w(l)
w(:\) =
(,!( II)-CC (1
I
I
I
I
-------i------I I I I
= O, this leads to a geornetric progression
w(2) = (1
I
I
I
a
CU 1
b
+ a + ... + aV-I) w(l)
Fig. 3.11. Starting a motor by successively reducing the starting resistor I.h,~ ~;1I11l
wit.ll
1,1(11)
I
Il
w(t)
v
w(l) = (1 -
I -- a
a
V )
(3.17)
Wo .
'1'11<' Li::L uI' Lhe N switching instants is reached when the natural torque/speed "IIIV" yJ('ld~ avalue of the torque which is less than the maxirnum, m2, (!(N) -
(1- aN)wo 2: (1-
m
2
mO n
lll(l/a) I"rl 1111 pl'(~viou~
N integer.
Wo ;
I, I ( I' )
I ,I
( , I (II
I I)
e-(t-t.)/T=(v)] ,
(3.19)
,
m2 e-(t-t.)/Tm(v) .
(3.20)
as soon as mM(t) has dropped to mI, the next contactor C v +1 is closed. The mechanical time constant, valid in the interval, is (3,18)
rcslllLs iL is lmown that, when assmning linear torque/spcecl ",II varia.hk:, arc ex pO!lclItinl f1lJlctio!ls oI' \.ime, w i th tl)(~ time COl1s\.all ts 1I"Il('llIlillJ', (III 1.1\1' ;'III!,II' ~; "r 1.11(: curve:; n(, I.!t(, poiul. (Ir i\l\.(:r: I( '.('.I.Íoll, III F Il'. :\.11 I I ::1.11111111', LrnJl:;i('Jll. wil.h"lIl. 1"11.11 h ill\(\wlI; III I,h,' iIlL('l'v;d
,'II r V( '1_)
w(v)] [1 -
where tv is the instant, when C v closes. The driving torque in the sarne interval follows also an exponential func tion
mM(t) = )
:;"lviul'; for N, this results in N /" 11l(mo n /7n2)
= w(v) + [wo -
T m (V ) --
J Wo _ J Wo v _ T a - mn mo(v) m2
mOn
m2
v,
a,
it decreases continuously as the starting resistor is reduced .
(3.21)
3.3 Numerical and Graphical Integration
3. Integration of the Simplified Equation of Motion
:\H
:L2 Analytical Solution
t2 = J
411' Nonlinear Differential Equation
/
= J Wo [ 2 mp
dw = mM(w) - mL(w) dt
J -
I 'y scparation and integration,
f () W2
t l -
tI
=J
-
o
fi direct solution of the equation
(3.22)
mM w 1-mL () w dw
Wl
is rarely possible because there is either no analytical expression of the func I.Íolls mM (w) and mL (w) or, if there is such a formula, the integral cannot be
J Wo /0 ( S =-Sp + -Sp) 2mp S
dw mM
W2
39
dS
52
1
-
S~
_ Sp IS] n 2
Sp
(3.23)
•
This function is plotted in Fig. 3.12 b. Since the speed approaches the syn chronous speed Wo asymptotically, the definition of a starting time requires the choice of a final steady state slip S2. Computing the starting transient on the basis of the steady state torque/ speed characteristic makes sense only if the starting is sufficiently slow in order to justify ignoring the electrical transients. This condition is met when the load contributes substantial additional inertia or when the start is delayed due to reduced volt age. Otherwise the start of a smaller motor would be completed within a few periods of the line voltages, long before the electrical transients have decayed; this will be discussed in Sect. 10.3.
so!ved in general terms.
ú)o
ú)o
1 c O ')p ", O,1 0,9
1
a
3.3 Numerical and Graphical Integration
..ill..
.!Q.
,
1 0,9
1
S
The most versatile method - with regard to accuracy and applicability of integrating nonlinear differential equations is the stepwise integration with the digital computer. It is known that by introducing state variables Xi (t), i = 1, 2, ... , n, a differential equation of n th order may be written as a system of n first order differential equations
j
+-t---_
dx ' d: = CPi(Xj, Yk, t),
m mp
b
Fil~'
3 .12. Starting of induction motor (11) Steady state torque/speed curve (b) Starting transient at no-load
Au exception is the no-Ioad starting of an induction motor without sta l.or rcsistance and skin effect in the rotor bars; the normalised steady state l.urque/speed function is
= S/Sp + Sp/S
'
1.111' i lll." )'.r·,,,\,l ,," ywlli ::
(3.24)
where Yk (t) are the independent forcing functions and t the time as vari able of integration. The functions CPi may be arbitrary, given as analytical expressions, curves or even discrete samples. Once the state variables Xi(t) are known, the output variables Zi(t) can be calculated from algebraic ex pressions, i.e. without further integration. Zi (t)
= cp i (x j, Yk , t) .
(3.25)
(v+l) Llt
wil.h S _ 1 - w/wo being the slip and Sp the pull-out slip at which maximum l.orq\l(~ 'l/1. p ocellfS. 'J'lw function which will be derived in Chapo 10 is seen in Fil'.. :~ , 12 a for SI' 0, I. Wil.h 110 \,,;ul ('111" . O) nnd tJw illil.in.! cOllllil.jolllv(t.l ' c.:. O) "-" 0,8(0)
,n,
'J'he simultaneous solution of the n state equations (3.24) is performed in s teps of length Llt
2mp rnM
i,j,k = 1, ...
=1
x;( (v
+ 1) Llt) =
Xi
(v Llt)
+
/
CPi( Xj , Yk ,
t) dt .
(3 .26)
v Llt
:/:; (/l I.\ t.)
;I:i (/I) are the results obtained in the preceding step, Xi(V + 1) is "r v lI.ll1cr; al. time (v+ 1) Llt. Da,.'i (·~ d 011 Hvproxirn ations provided hy 1II III.h"III i\ l.i ni alld wil.iI Hullici('ut.ly Hlll a ll :; I,I:p h:1l ~ l.h wc CHIl wril.l~
I.h (· IU'W ~1(' 1. 11I1I1I( ' I' k rtl
,III
3. Integration or the Simplified Equation or Motion
:Ci (V
+ 1) ~
Xi(V)
+ F['Pi(V),
'Pi(V - 1), ... ] ,
3.3 Numerical and Graphical Integration
(3.27)
w!ln"!! F is a linear combination of the functions 'Pi at previous intervals. 'I'Itere exists a large number of integration formulae which differ in complex il,y, i.e. computing time, sensitivity to numerical instability and accuracy [22]. 'l'lte ones best known are square-, trapezoidal- and Simpsons-rules, further IIl
dv
1
-1--
=M
ds dt
= V,
(t
[JM(V, t) - h(s, v, t)] ,
wll(!l'(' 1M is the internal driving force of the locomotive and h is the to tal loar! force, íncluding frictional and braking forces a.s well as gravitational l'ol'( 'cs Oll grades. The task of finding an acceptaiJl(' tillldable is an optimisa I,i(lll prohlclJI of cOIlsiderable complexity, sinc(' lIlally cOllstraiJlts and bound :lr ,Y cOlldi(,ioIlS llIust h( ! tõlken into aCCOllnl.. 'l'llú; indlld,':; Lhe Il S(! of the same Lr:lt'k I,y I.r:tills Itavill g diffcrc llt I\.('cdcra(.ioll :1,lId vl'lociLy lilllÍts, such a.'i in L('IT,iLy
41
Since it cannot be ruled out that, even today, this may have to be done occasionally, let us briefly consider the principIe of graphical integration, using a simple example. Eq. (3.1), where mM(w) and mdw) are given as empirical curves, Fig.3.13, J
dw
di = mM(w) -
mdw)
= ma(w);
(3.28)
the equation is first normalised with the help of arbitrary values J WI ~ (~) dt
mI
WI
mM _ mL _
_ -
mI
mI
-
ma
WI,
mI, (3.29)
mI '
where ma/mI corresponds to the normalised accelerating torque.
cu
mL
/
/
/
/
/
m Fig. 3.13. Torque/speed curves or motor and load
With the abbreviations JWI mI
= TI,
t
-T=r 1
W
'
WI
= X,
ma mI
=y
a nondimensional equation results
dx
dr =y(x).
(3.30)
Normalisation avoids the awkward choice of scale factors; the reference values are of no significance, even though it is recommendable to choose characteristic values such as nominal speed and torque in order to work with handy numbers . ln Fig. 3.14 the normalised curve y(x), obtained from Fig. 3.13 is shown, with t.lt p lI ol:llInlif;cd speed plotted against Ilormalised torque. The x- axis is sllbdividcd illl.o scv( ~ r;tl illt(!l'vals, JlOt, !)(!cPHsarily of c<.J. l1al lcngth; it is in fact :l.pPl'Opri:i.I,,' 1.0 /'C'dll('" 1,111' 1(!lIgLhH 01' 1. Ili' illkrv:d:; iII /;cdiolJ S wheJ'c y(:,:) iH ('Ia ii "1',1 " I', I'lI. lIidl y , 111":,,1'11 illl.(·rV:I,1 y i: : ' i.J'I" 'II ' "II :tf,(·cJ h,Y II, (·IIII I
I\~
3. Integration of the Simplified Equation of Motion
x=
4. Thermal Effects in Electrical Machines
01 01 1
x -----.......... .......
t
T =r
1
ma
y= m
r
1
Fig. 3.14. PrincipIe of graphical integration
rCJlresenting the average of y(x) in that interval. Approximating y(x) by a function has the consequence that dx j dr is assumed constant in ('
~it.airc ase
4.1 Power Losses and Temperature Restrictions 80 far our considerations were only dealing with mechanical phenomena and the pertinent steady-state and dynamic conditions, but suitable torquejspeed curves and adequate power are not the only criteria for designing electrical drives. Of equal importance are the thermal transients in the motor caused by the unavoidable power losses during the process of energy conversion and the ensuing heat flow from the points of origin to the cooling medium. The various materiais used in an electrical machine naturally have different temperature limits; of particular importance are the insulating materiais, which determine the temperature classes of the machine, for example class A, B, F, H,
Lh9 Lh9 Lh9 Lh9
< 60° C, < 80° C, < 105° C, < 125° C,
cotton, synthetic, paper, resin, shellac, mica, epoxy, glass fibre, silicone rubber.
..119 is the mean temperature rise above an assumed ambient temperature of 190 = 40°C. All temperatures are measured in degree Celsius (centigrades). For a given frame size and type of cooling, the output power and the allowable power losses are higher for the elevated temperature insulation classes; at the sarne time the cost of the machine increases. The main causes of power losses in an electrical machine are: a) Conductor heat loss (copper losses) in the windings, cables, brushes, slip rings and commutator. If there is alternating current, the effective resis tance of the conductor may be noticeably increased by eddy currents (skin efFect); this is accentuated when large conductors are surrounded by mag netic material and may have to be alleviated by the use of stranded con ductors . h) Irou heat losses in stationary or moving magnetic material exposed to dl
"r
t.llI' Itl n /',rrdi .. lidd .
,1,1
4.2 Heating or a Homogeneous Body
4. Thermal Effects in Electrical Machines
I' ) I,~,.ú;tion
'I'he various types of losses depend in complicated fashion on the oper :d.iollal state of the machine. The main factors of influence are speed and lo:ul torque, as well as voltages and currents with their RMS-values and w:wdorms. 111 order to induce heat flow from the points of origin to the cooling ::llrf'aces, a temperature gradient arises inside the machine determining the
[
lasses, including bearing-, brush- and ventilation losses.
= L119
+ (Ll-!9 max
L119) + {)o
40°C = 130°C.
-
~IA
Pj
t)
C
o ~P2
t}O Fig. 4.1. Homogeneous body with thermal storage capacity
power Pl, at the same time emitting heat power pz by convection. The am bient temperature is {)o, the coefficient of heat transfer a. The component of radiated heat is assumed to be negligible due to the relatively low operating temperatures (it rises with the fourth power of absolute temperature) and due to back radiation, for example at the cooling fins. Hence the equation, describing the power balance is d{)
C- =Pl -P2. WlwlI operating the machine at elevated ambient temperatures, e.g. in the I.ropies, the power rating may have to be reduced. A n accurate prediction of the heat flow and the temperature distribution iII :w dect.rical machine is exceedingly difficult; this is due to the complex I';( 'ollldrical shapes, the use of heterogeneous and anisotropic materiaIs (lam ill:J!,I'd irou , insulation), furthermore the complicated laws ofheat generation wiLh n'sl)(~d to space and time, as well as the different cooling conditions. A b " Lhe l)(~al. conductivity of the various materiaIs does not differ by orders II/" 1I1:J.1',lIiLlldc', as is the case with electrical or magnetic fields. As a result, ti ... 1."II'IWra(.llre distribution can only be computed in restricted sections and wli,1I (·oll.'ii(krable simplifications. TI", IISI~r (Ir dectrical drives has normally little influence on the tempera 1.111'" di :-;(.rihlll.ioll within the machine. He must be confident that the designer II /lS dlllSI'1I slIHiciently large conductors and cooling channels so as not to I· F I"'.! LClllpl~rature limits under specified operating conditions. With this IU il'i lllll(.Lioll iI. is usually adequate to drastically simplify the thermal model ,,( Lili'. "lllOlor" by regarding it as a homogeneous body exhibiting thermal : i L()ril-f.'.I~, whose internal transients are unknown and of no interest. Naturally, ::<,,'11 a I:rude rnodel cannot offer any detailed information on specific internal I.I11'I'IIlal I ~ollditions.
LIeating of a Homogeneous Body
'I'I,,~ :;illlplifi(~d l, hl ~ J'lllallII()dd
is
charactl'~ris('d
hy l.hc followiug aSSlIlllptiolls,
I"i'r', . ,I. I ;
fi t.lIlI'd
1""lv wil.lt tllI' :illr(",("I ' 11, 1.1 ... I'.I11'1'1 II ii. I (':Ip:u 'il'.y (), 1111':1. \lV II ' C, 1111..1 1.1 .. ' 1111'11.11 ::111 /'lI,(' I' 1."1111'('.,," 1,111" ,'I I; : 1.... \.1"'1 I"v I.hl ' illjlllL
hll I IIOI 'l "lIf'IIII II iII
(4.1)
dt
The heat transfer by convection is a linear function of the temperature dif ference
pz
=
with {) -
a A ({) - {)o) ; {)o
d(L1{))
=
C~
L1{)
+a
(4.2)
this results in
A L1{)
= Pl
(4.3)
or
T dL1{) {) dt
+
L1{)
= ~ a A '
(4.4)
a linear first order differential equation; T{) = C/ aA is the thermal time constant. With Pl = PlO = consto and L1{)(O) = O the solution is
L1{)(t) = L1{)(oo) (1 - e-t/T~) ,
(4.5)
where
L1{)(oo) = PlO/a A f1.~
45
(4.6)
is the steady state end temperature, at which the heat transfer by convection equals the input power. lu Fir;. 4.2 a tlwrrnal transient is seen. The shaded area between pI/a A aud Lh'l(t,) c()ITI',~p()llds to the energy stored as heat. Hence, the storage effect 1.1'11(1:: 1.11 dc'I:I,)' 1.11f' 1 . l'llll)(~l'at.lIn! chaJlg(!:;, ea.llsiJlp; the temperature to be a (""lIl.illllllll~: 1'11111'1.;1111 III' 1.;1111',
4. Thermal Effects in Electrical Machines
~ ti
\ I~
LW(oo)
4.2 Heating of a Homogeneous Body
little thermal storage capacity. The following time scales are usually valid with electrical machines:
L1~
= P10 aA
L1~(O)
electromagnetic 0.1 - 100 ms
TJJ
TJJ
cooling-off transient
a= 50t0500WtCm
Fig. 4.2. Thermal transients
The approximation incurred with this simple model is mainly due to the assumption of evenly distributed heat sources in the homogeneous body and t\w Ileglect of an internal heat flow; hence, transportation delays are not rOlltained in the simplified model. A rough estimate of a typical thermallag may be derived from the data sll('d. of an enclosed 100 kW standard induction motor cooled by an external \';\.1I (TEFC): 100 k W,
) ' /I
nominal power mass 0.92, nominal efliciency GO°C is the steady-state temperature rise due to the losses PI at
~ OO
M ,I . I,,') (' x,)
kg,
11
1'"
1'1
(~-1)
LIli' C(l lI v( ' cl.ive
11.
-:"c
that the motor consists of solid iron having the specific heat capacity is
~ =
A
j
2
= 2.5 to 25 kW m ;
the dependence of the heat transfer on the velo city of the cooling air has the consequence that self-cooled motors exhibit at standstill a cooling time constant that is much longer than the thermal time constant of the motor when running. Larger motors operating at variable speed and load are usually provided with separate forced cooling, causing the motor speed to have little effect on the cooling conditions. To show that the heat transfer by radiation can normally be neglected, a rough estimate of the power flow by radiation is given for comparison. A "black body" emits to the environment power by radiation according to Boltzmanns law,
= uT 4
,
u = 5.710- 8 W jK 4 m 2 CFe,
'I'his yidds the thermal time constant
;'}
2
where
L119(oo)
c = Cl"e M. "
P2c
A
J\ h S lllllill~ I. !tI , hea.L
thermal transients 10 - 60 min
at Ll79 = 50°C this results in a power flow by convection of
PR
heat conductivity is
PI
Ir
mechanical 10 ms - 10 s
Travelling wave phenomena in windings are not considered; they are faster yet by several orders of magnitude. The large difference of more than a factor of 100 between the time scales of mechanical and thermal transients usually permits a separate treatment which results in considerable simplification. The heat transfer by convection from a ventilated surface depends strongly on the velocity of the cooling air; the usual range of the transfer coefficient is
Change of stored thermal energy
hea ting transient
47
C/"e
M Ll19(oo)
=
.!.::.'L p 11.. 71
n 11.
'l'y pirall y , Lhcl'lllal
Lrll.llsj('lI(.s ar(:
1'1\', '1'1.::
()I' ((1111 ::" ,
1.111'\'"
111 :.1. 111"
Wl ll d i JlI',
0.48 kWs j kgOC 800 kg 50°C ~ 34 mino 9 kW
IIl1lch I;!ow(:r I.hall d( ,ct. rj(·:t! 01' J11(·('hallica.! di/l l( 1II(1(.tll·H, wlll'l'! ' 1.111' nr Irl I' Ji ui<'d d i ,.,·(·l.Iy (III !III I lI li lllll,I,I II P, d lll lt I UIfI IJl IIII H'III II ':I V('I 'y 11.1 '(' (·:x('(:p\.j(lIl H :1111'11 III '
is the coefficient ofradiation and T the absolute temperature. With 19 = 70°C, i.e. T = 343 K, the power flow due to radiation is PR
A
= 0.77kWjm 2 .
This relatively small amount is further reduced by inverse radiation. AIso, the r acliation coefficieIlt of a motor surface is much lower than that of a "black hod y" .
is the extrapolated end temperature during overload. Hence
4.3 Different Modes of Operation
Pls
III contrast to internal combustion engines, most electrical drives possess (", Hlsiderable overload-capacity, the limit of which is determined by the power losses rising with the load. Since the temperature follows the power loss with :I. slIhstantiallag, a temporary overload is admissible without exceeding the 1."llIperature limits, provided that the motor is initially below the nominal 1.' ~ llIperature. This effect may be utilised when selecting a drive for a particular :tpplication. Let us look at some modes of operation frequently occurring in jll'actice. il.:~.l
49
4.3 Different Modes of Operation
4. Thermal Effects in Electrical Machines
I\{\
-
Pln
::; 1
-
1
T{)
-t
-t '
e 1 IT~ ~
1
t1
« T{) ,
(4.9)
is the maximum relative power loss during the time tI. lncreasing the output power while simultaneously reducing the time h of operation is of course only feasible within the available torque range; this is illustrated in Fig. 4.4. At the left, a typical curve relating torque to power loss is drawn, while the other curve depicts Eq. (4.9). This indicates that the power loss rises much faster than the torque (assuming constant speed). The shaded contour corresponds to the admissible short time overload range.
Continuous Duty PIS
TI! is lIIode of operation exists if the load torque is constant during an ex LCllIl(~d period of time, corresponding to a multiple of the thermal time con
SL;\lIt, so that the temperature reaches its steady- state value, Fig. 4.2 a. 'l'his is the normal continuous duty, e.g. with paper mill drives or boiler feed p11111p5. The motor must be so chosen that its nominal output power equals "I' ('xceeds the continuous load. With traction drives, other definitions are also ill use, such as a one-hour- or five-minutes-rating of the drive motors. tl. :.L 2 Short Time Intermittent Duty
III t.hi s
Lhe tillle of operation is considerably less than the thermal time :wc! t1w motor is allowed to cool off before a new load cycle be 1'111::, 'l'1,i ~i dllLy apilhes for example to some crane drives or small motors in j"'II::,'III>I
I i (,i()lIS. ln Fig. 4.3 the temperature transient is drawn, following a short time pmv,:r lo:{s 1'1. which may be considerably in excess of the nominal power lo:;s 111",' The temperature rises rapidly towards an extrapolated steady-state vai 11(' L119 j when L119 1 , is reached, the motor is disconnected, giving it the 00 npport.lluity to cool. ln the example it is assumed that the motor is self mole!! C
, ' , 1111
V;t.!IIC ,
Tlw Illaximum t.cmperature L119 1 , at the time of disconnection is
A" ~' I1 Q()
[1
-
<
C - t';T~] _
A" .L..l'f!rna :c ,
(4.7)
wlwrl' I
\n ,. ,
I
\1'/
IIJ" , 111 "
1I I I I
I
(J! .r\)
,119 - PIO
! ~ - aA
I I I I I I
-4~rJ-t-: o
o
, ' , II 1: , !.;utL
A 'I I w'U
,119
ti
TjJ
Fig. 4.3. Thermal transient during short time duty
m
ti
mn
Ti}
Fig. 4.4. Short time overload limits
With motors exhibiting a pull-out effect there is an absolute torque limit, whereas the losses continue to rise up to the stalled condition; on commutator machines, the commutator is usually the weakest link. To realise high torque pulses which exceed even the short time overload capacity described by Fig. 4.4, a mechanical energy storage device, for in stance a fiywheel, may be used; of course, the inertia must be accelerated to sufficient speed before the load can be applied. This is the usual solution with highly pulsating loads such as punching presses. 4.3.3 Periodic intermittent duty
With some types of load, for example elevators, automatic machine tools, mine hoists or rolling mill drives typicalload cycles exist which are repeated periodiea.lly. A very simpie case is seen in Fig. 4.5, where the power loss Pli follow s a rectallGular periodic pattern. This n:slllLs iII t.( ~ lll])('rature fluctuation s, tlte rnean of which rises until a :' I.(·ady :.;1,;1.1,( ' ("')\ldiLioll iI; l"(·ac1l<'d. Wit.lt !.lI< : (,IWl'lll;d I',Ínw ("on stanls TI 7'~ valid dllrill/" (hl (11111 ()I!' l,iIlH' S /' 1, t 'l. , !.II(' rollowi ll l', rd:t.l.ioll s Itnld: I
~i()
4. Thermal Effects in Electrical Machines
T1
,1t'}+
i
I.
,1t'}
1-
00 . ...
5. Separately Excited
I .1
l --...
T2
De Machine
.1
I
---l
I
1
~~(1) i
1,11J(2)
I
/ //
//+-
/
/Ll1J'(1)
: -- ;1t'}'(2) I
I
t1
o
2(t 1+t2)
t 1+t2
5.1 Introduction
Fig. 4.5. Periodic intermittent duty cycle
L1t9'(v + 1) = L1t9(v) e- t2 / T2 , L1t9(v + 1) = L1t9'(v + 1) e-tl/Tl l';liminating L119' (v
L1t9(v
+ 1)
+ 1)
= L1t9(0)
(4.10)
+ L1t9 oo [l -
e-tI/Tl] .
yields
+ a L1t9(v)
(4.11)
,
lN lt( ~ lT
(4.12)
L1'/?(O) = [1 -
e-tl/Tl] L1t9 oo . a = e-(tI/Tl +t2 / T 2) 1.
<
(1.11) represents a linear recursion between two successive peak 01' l.lw tcmperature. The solution is a geometrical progression
1':qll:tI,ioll Vío.lll. '~:
,
/\/1(11) =
1- a IJ +I
l-a
L1t9(0)
=
1 - e-h / Tl 1 _ e-(tI/Tl+t2 / T 2) [1- e-(IJ+I)(tl / T l +t 2/T2)j L119 00 .
+ v (h a time constant
I kC;lllse of t = tI liavillg
'L'
. ,'I
~
tl
t2 ITh1 + + t2 IT2
+ t2) .
wlt
(4.13)
the peak values lie on an exponential function
h
TI < T{) < T 2·
(4.14)
'I'IIC st.(~a.dy state peak temperature follows from Eq. (4.13)
Ll1~( (0) = Wit.h
1_
1 - e-h/Tl
rorc<~d v"IlLilal.ioll ,
'1',
'{'.
'('i I
(4.15)
(~- (td'J\+t2/T2\ L1'Ooo . n sJlccial
('.as(~
(' X
ist.:; wil,h
Direct current (DC) motors have been dominating the field of adjustable speed drives for over a centuryj they are still the most common choice if a controlled electrical drive operating over a wide speed range is specified. This is due to their excellent operational properties and control characteristicsj the only essential disadvantage is the mechanical commutator which restricts the power and speed of the motor, increases the inertia and the axiallength and requires periodic maintenance. With alternating current (AC) motors, fed by variable frequency static power converters, the commutator is eliminated, however at the cost of increased complexity of control. This is one of the reasons why controlled AC drives could not immediately supplant DC drives, once the semiconductor-technology had sufficiently advanced. The principie of a DC machine operating in steady state is assumed to be known [4,64], but let us recall some basic facts. ln Fig. 5.1a a schematic cross-section of a two-pole DC machine is shown, containing the fixed stator S and the cylindrical rotor, called armature A. While rotor and pole shoes are always laminated to reduce the iron los ses caused by the varying magnetic field, the rest of the stator is laminated only in larger machines, when the motor is required to operate with rapidly varying torque and sp eed or when a static power converter with highly distorted voltages and currents is employed as the power supply. The main poles MP are fitted with the field windings, carrying the field current ie which drives the main flux Pe through stator and rotor. A closed armature winding is placed in the axial slots of the rotor and connected to the commutator barsj it is supplied through the brushes and the commutator with the armature current ia. This creates a distributed ampereturns (m.m.f.) wave, fixed in space and orientated in the direction of the quadrature axis, orthogonal to the main field axis, so that maximum torque for a given armature current is produced. ln view of the large airgap in the quadrature direction the resulting ar mature flux P" is much smaller than the main flux Pe. It can be reduced CV(~\l furtll er hy placing compensating windings CW in axial slots on the pote sh()( ~s Hlld ( ' ()lIlJ(~dillg "Item iu ::led es with the armature. Their opposing 11,1111)('1,\,1 ,111'11 1-1 11 11 1'111.
(, ;\.IlI'd
til<'
qlladral,lIl'< '
li"'d ('x eiL(,,1 hy
I.IJ( ~ a rtnal,lIlT
u.II
5.1 Introduetion
5. Separately Exeited DC Maehine
;,2
ar ·
QUadratture axlS
Neutral Lone
I I I I J J J J
JJ
~ff /
/ /
ie
.y~// / V
/
/
/
/
ia{t)t fii t+M )
iit)
ia , .
b
i 1/
Ua
/
/
/
a
a
ar ·
I I I I I J J
G
ie
li
Commutating coils
Direct axis
ia
a
53
:
tf
+
Commutator
iit+M )
Current in
ne7gnoourTng p--- conductor I:
Jo'i~~. G.lo (a) Cross-seetion and (b) sehematie eireuit of a DC maehine
b
----'VrCIIIOV(: t.he undesired armature reaction, which otherwise tends to distort the I'V('II di,~t.ribution of the main field under the poles along the circumference (Ir !.II(' rot.or. Compensating windings are common only on larger machines 1>1 ' ('()llvcrt.(:r-fcd motors for heavy duty applications such as traction or steel Illdl driv(:s . Compensated DC machines can withstand higher overloads than IlIJ('UIII\)(:ll satc(1 machinesj also the armature current may rise much faster 11.11< I larl~( T current harmonics are acceptable without detrimental effects on l'Ollllllllt.at.ioll, i. e. sparking of brushes. This is of particular importance if the 111:\.('i1iIlC is supplied by a static converter. 'I'!t (: co lI1mutating poles CP, placed between the main poles and also rarryillg t.hc a rm ature current, have the task of locally modifying the field in I.h(' ll Cllt.ral zone, to achieve rapid and sparkfree commutation. This is done I,,Y illdll Cillg a suitable voltage in the armature coil tempora rily shorted by 1.lte hrllshes. ' l'lw priuciple of commutation is illustrated in Fig. 5.2, where the c10sed n l'lll a l.lln' wiuding of a two- pole DC motor is schCIllatically drawn , showing I.hl ' hrllsll j)osit.iulls a t two cO!lseclltive ius tallt.H 0(' t.illl<:. Clcarly, when the r( 'c
--!?- - - - - J :
/l.
Current in an
armature conductor
Fig. 5.2. Commutation of eurrent in an armature eoil, (a) Winding and (b) Cur rents in two adjaeent armature eonduetors
commutator acts like a multi-phase electromechanical switching converter which is synchronised by the rotor angle. While the current is being inverted, the commutating coil is temporarily short-circuited by the brushes overlaping adjacent commutator bars. Since each coil, embedded in slots and surrounded by iron , exhibits some induc tance, the commutation is a continuous process which requires a finite timej this limits the speed at which the motor can operate without excessively sparking brushes. For a two pole motor with 100 commutator bars operating at 3000 rev /min, the overlap takes about 200 f..Ls. If, due to excessive speed , the short circuit of the commutating coil is opened before the commutation is completed, a voltage is induced which causes sparking at the trailing edges of the brushes. TlI c) e[pc:\.rical torque mM is exerted at the rotor surface; more precisely, hy t,a.ll ~(~lIt.i a) llJagllctic forces acting on the sides of the slots. It is propor t.iOllil.I Ln t. IJ( ~ prodlict. 01' Illa.in f1l1 x and ill'llla(;lIrc CllITCnt. The volt.ages in the 1l.1'I11 i1.l.lliT ('i r('lIil. (·t1l1 siKt. 01' ali ill
\
5. Separately Excited
~ ', Ij
De Machine
5.2 Mathematical Model of the
IIllx and speed, in addition to resistive and inductive volt age drops in the commutating- and compensating windings as well as the arma 1. 11 r< ~ connections and brushes. The main flux is thus of prime importance for opcrating and controlling the De machine; greatest flexibility exists, when Lhe field current is supplied by an independent source so that it can be varied ;d, will. ln Abb. 5.1 b a simplified scheme of a separately excited De machine is shown, where the resulting magnetic fluxes are concentrated into lumped illclllctances which are orientated in parallel and orthogonal to the main field a x is. The total armature volt age Uo. thus consists of the volt age e, induced hy rotation, the voltage of self-induction Lo. dio./ dt and the ohmic drop Raia, whcre La, ROo are the resulting parameters in the armature circuito With la,rgcr machines, Raio. « Uo. is valid at normal speed and load. AIso, the p()wcr required for excitation, i.e. the losses in the field winding, amount to nlll y a few percent of the power converted in the armature. With motors having permanent magnet excitation the power losses in the lidd winding and the need for a separate field power supply are avoided at Lhe cost of operational flexibility. Servomotors for machine tool feed drives ;m! llsually of the permanent magnet type. Thc De machine as a dynamic system, including the interactions of the d(!('tromagnetic and the mechanical effects, is dealt with in the next section.
ia
; ~rlllature-,
;1.:2 Mathematical Model of the
De
Machine
'1'11<- (!quivalcnt circuit of the separately excited De machine can be repre
schematic form by Fig. 5.3 containing concentrated components 11.11<1 :i l!owillg no longer a geometrical resemblance to the actual system. Til(! lH!rtinent differential equations are
::( ' 111.('<1 iII
Il" ('
'111M
=
=
,
'I".
'}', ti,
tlt
La ~ ctt + e = Uo.
-I-
Newtons law, concentrated mass
mM - mL ('2
at
N d
electrical torque
= Ue
field circuit
J(i e}
static magnetising curve, omitting hysteresis angular velo city.
=w 1 ':lilllillaLill/~
("i
armature circuit
La
ueQ
Ua
55
eJ J
Fig. 5.3. Equivalent circuit of a separately excited
De machine
where e ia is the instantaneous electrical power converted in the armature to the mechanical power mM W. Both must be identical, hence CI = C2 = C. Normalisation with the nominal values Wo
= eo = c
20.0
= ROo
mo UeO
= c
nominal speed at no-Ioad (base speed), nominal flux nominal armature voltage extr. current at standstill with nominal voltage (8-10 times nominal current)
extrapolated torque at standstill
nominal field voltage
results in the following non-dimensional state equations T
d o. dt
d T eo dt
(ii) _Uo. io.o - Uo.O -
ia W
~-fe (~:) = UeO (~:)
,
Lo. To. = ROo;
T
eo
= N e
(5.1)
(5.2)
induced volt age (e.m.f.)
P" w
/:1
llw ,j Ilt
}l"
I
'I,,,
Ra
De Machine
"I
H
'J
III ~ I I "
froul I
t.h(~
S('<:()\1<1 H.u
Tm
:t (:) (ê) =
d TE; dt
êO
ia
w
Wo
mL
mo
T _ êo E; - Wo
T
_ Jwo • mO '
m -
(5.3)
(5.4)
ie/ieo = fe(
5.3 Steady State Characteristics with Armature and Field Control
S. Separately Excited DC Machine
!dl
=0, ( ~) PeO .
~-fe
mL3
mo
UeO
57
(5.7)
mu mo~
~~
_ mL
PeO
mo
iaO E
'.!iL.
Eo
" ,' ,~
(5.8)
The angle of rotation e changes linearly with time; it is of no consequence for the time being. For the following discussion the normalised flux b=
~<1
PeO
which is limited by saturation of the iron core, is assumed as an independent control input, instead of ue/ueO. This results in
11;,0
Ue
Ueo
Jo'i~.
=0.
W _
Wo
ia
5.4. Block diagram of a separately excited DC machine
iao
L" i.ll( ~
larger airgap in the quadrature axis and the possible presence of a com 1H'll sal;ing winding Ta « T eo holds. Magnetic saturation in the quadrature ti I !('ctiO!l is usually negligi ble. Tb(~ dynamic system, described by Eqs. (5.1)-(5.4), is represented as a hlock diagram in Fig. 5.4, containing linear integrators with their time con ,~L;wts and nonlinear algebraic functional blocks; it is assumed that the load Loj'(l'w consists of an independent component mLl acting as a disturbance ;1.11<1 lWI) components as nonlinear functions of speed and position respec I.ivdy; wlwn the load torque shows no such dependence, the corresponding 1'1 I II cl.i OIlS are zero.
. - - - + f L2 ( -W ) + f L3
'/II./, ._ nlLl ~ W'll
mo
Wo
(e) eo
.
(5.5)
1
1
Ua
- b UaO
-
mL
b2 mo '
= ~ mL b mo
(5.9)
(5.10)
Hence the steady state behaviour of the controlled motor,
~=h
Ua mL) ( uao' mo
~=h
(:~)
Wo
iao
'
is characterised by linear functions, connecting speed, torque and armature current. Depending on whether U a or Pe , i.e. Ue, serves as control input, the term armature- or field-control applies. 5.3.1 Armature ControI
;•. :~ Steady State Characteristics wil.h Armature and Field Control 'I'lw
~t(~:tdy :;t,
i,; ()hl,;ÜII(~d
Assuming b = Pe/PeO = 1, the nonlinear effect of the two multiplications in I"ig. 5.4 vanishes and a particularly sim pIe set of linear control curves results,
cOIldiLiol\ , valid with constant input quantities ?La, u(., hy sdLilll~ til(' d( ~rivalive~ iu Eqs. (5.1)- (5.3) ('<juallo 'l;ero,
'U 4 ,
'/.4.
UI/I)
1 .. 11
( tJ
,I'.
{"u 1/'1'11
II,
'In/,
u,·(')
W
Ua
mL
Wo
'Uao
mo
'I.'L
UI. /.
"/II)
'III.()
(5.11)
(5.12)
r.H
5.3 Steady State Characteristics with Armature and Field ControI
5. Separately Excited DC Machine
.!L
(j)
1 1 1
iaO
59
the motor designer would not have fulIy utilised the potential of the motor hence it is only feasible to weaken the field,
1
-1 ~ b = ip e / ip eO ~ 1 .
1 1 1
10,5 1 1
I
-1
Normal operating range
a Fi~.
b
:,illcc th c armature voltage U a is referred to its nominal value Uao, only t.I)(' Jalll~(~ -1 ~ ua/uaO ~ 1 is of interest. Excessive voltage would cause :11'11.1"11 i 11!', of t.he brushes and possibly commutation failure by formation of an
::iJ()rt.· cin:uiting the brushes (flashover).
'l'lw al"luature current is proportional to the torque, independent of volt ' 1) \" :1.1141 s p(~ed. Armature current and torque are normalised with their ex I.ra» olal.cd values at standstill. Therefore the normal operating region in Abb. ~I.~J I, is (:olltained in a narrow band, e.g.
-0.2
7nL
<-
rno
i = ~ ~ 0.2 , ZaO
whcll allowing a temporary overload of about twice nominal torque. ()liI.:, ide t,his range, the curves are likely to be distorted by armature reaction IUld I.h c !"c ll1ay be commutation problems, part.i<:lllar1y ou lllachines without ;1. ("oll1pt:ll~;al.illg winding.
, 'V,' II
. J. :I. ~
Fi(' ld COlll.rol
mL
standstill:
W
= O,
WNL
Wo
I )n;v
(,\I
1llll. llrl d,LlIlI 'I',.
(· llllllol.
I,,·
I, y ,'1. /1111'.111/ ; Ir,lI e IJl1I.iu flu x 111111 1'1 1 1" .' V' '11d 'I', fi "1.",,rw i ~I( ·
IU"(." " \"
'·I U' I" ' II t1(·d,
= O,
mst
mo
=~
b'
=b .
With reduced flux , the no-Ioad speed rises, while the (extrapolated) stalIed torque is reduced. Hence the slope of the curve is varying with 1/b2 ; this inherently "soft" load behaviour is undesirable with drives to be controlIed for constant speed. As is seen in Fig. 5.6 b, the armature current and the armature losses in crease when the motor is operated at a given torque with reduced field; there fore, as already mentioned, field weakening makes only sense if the desired operating point in the torque-speed plane cannot be reached by armature voltage control at fulI field or if tbe armature voltage is fixed. At less than maximum armature voltage, operation at reduced field is uncommon, except in special circumstances, for instance when several motors are supplied from a common armature bus voltage and only minor speed variations are needed as on some paper mill drive systems. Neither is field weakening a solution for the start-up of motors since at standstill the armature current is not affected by the main fluxo Fig. 5.6 a indicates also that a reduction of the field does not necessari1y lead to higher speed; because of the steeper torque/speed curves the result of field-weakening may even be reduced speed at higher armature current which is of course most undesirable. The maximum speed at a given Lorque is obtained by differentiating Eq. (5.9) with the condition U a = UaO,
ri W'" W
Til" 14" \""[[4 1 ''111.1011 fo r .." n Ll",,[,jill 'll I.hll
no load:
5.5. Steady state curves of DC motor with armature contrai, Pc = PeO
!lu:y
11.r ..
If the armature power supply permits operation in alI four quadrants of the ia-plane, it is sufficient to restrict field control to positive values, bmín < b ~ l. The purpose of field weakening is to raise the speed at light load, as seen from Eq. (5.9), accepting the disadvantage of increasing the armature current for a given torque, Eq. (5.10). Therefore it is normally not appropriate to employ field weakening unless the possibilities of armature control have already been fulIy exploited. Inserting ua/uaO = ±1 into Eq. (5.9) yields a new set of steady state curves with the normalised field factor b as control parameter; the character istics are again straight lines, however with increased slope (Fig. 5.6). The speed/torque curves intersect with the axes at
Ua,
tlll
I ,'/tI ,
= _ _1 ('()ll"( ' ,) ,
whi .. h 1" 1\.,[ ,1 1. ..
II.
b2
+ _2
mL
= O
b3 mo
p rllcl.i cnl lowcr Iilllil
1'0)'
Lh (~
fit:ld
1;11
5. Separately Excited DC Machine
(U
5.3 Steady State Characteristics with Armature and Field Control
..!L
(f )O
torque but high rolling speeds are desired for saving time. A similar situa tion exists with coilers where high speed and low torque are needed when the diameter of the coil is small, while the reverse is true when a large coil diameter has built up. After eliminating the field factor b from Eqs. (5.9), (5.10) with the con dition u a /uaO = I, an expression for the mechanical power is derived
iaO
.7
1
1
0,5
w mL
Wo mo
!!!..L
1
mo
!!!..L mo
operating range
a
b
Fil-'; . 5.6. Steady- state curves of DC motor in field control range, mL
--, I',",i" -- 2 mo
Ua
= UaO (5.13)
w lll'J'(' Lhe undesirable speed reduction begins. Note that mo is the ex f.liI.polak
I !j
'/1/.0 I
II 'I I L I,
(5.14)
", I,i,·11 1"'IH'('SCllts ali envelope for the family of torque/speed-curves with pa I ; t.lII'·1,1"I' I, . lVIaximum power is obtained at the point, where the straight line i:: 1. :1.1'1';"111. 1.0 the hyperbola. To avoid the situation mentioned before, the
sl!oltld lIever be operated at the right hand si de of the tangent point.
11101,01
~
IIlachillcs without compensating windings show a pronounced arma in the field weakening range because the field distribution in 1.1 1<' a i I'f ';ap i,~ 110 longer stabilised by saturation. Therefore the validity of the l.orqlll:/sl)(!('d -cnrves is restricted to reduced values of the normalised arma 1111'(' CIIIT(!lll. ; higher currents have detrimental effects on commutation and 1I1:1,,\' ('('slIlt iII sparking of the brushes. This calls for a rcduction of the max illllllll :i)'] II ature current at reduced field and hente a [low(!r derating. I )c::piL(! tlt esc drawbacks, field wcalccnill[.\ is n valoald(! lllcallS for increas i 11/', UI/: sl )('(;d ill t.Jw low torque r egiOllj typi(:al ll.pplil"it.l,io ll !-l an: spilldle drives illl Jllitchill(; l.oo1 s to allow cl\t.til\l~ at, fllll pOlVt'I , i'Vi'lI wil.lt a slllal! diallH'Ler W Clrli pi ( '( ' (' III' I.oo!. Ot.!w(' ('X illllpks lU< ' rt'Vi'), N h lj ~ I "I II II/', IIIIII " .!"III' l'o
1111'"
("( ;a("(.i O Il
61
ia iao
(1-~)
(5.15)
iao'
which only depends on the normalised armature current. Rence, with lim ited armature current, the motor can be operated at variable speed with a given maximum power , indicating that field weakening has a similar effect as a variable gear for increasing the speed of the load. For the reasons al ready mentioned and because of the mechanical stresses, the maximum speed achieved by field weakening is sei dom higher than two or three times base speed depending on the design and the size of the machine. Rowever, there are exceptions, such as on vehicle or servo drives, where a much higher speed ratio may be necessary. The similarity of a speed increase by field control and by a variable gear is corroborated by their effects on the mechanical time constant. At nominal field, Pe = Peo, the nominal time constant is Tm = Jwo/mo, where Wo is the base (no load) speed and mo the extrapolated stalled torque at nominal armature voltage and field . With reduced field, the no load speed is raised to b-lwo , while the stalled torque is lowered to b mo. Rence the mechanical time constant is a quadratic function of the field factor,
Tm (b ~ 1) = J Wo b- 2 == Tm b- 2
•
mo
A similar effect is observed if the load is coupled to the motor through a step-up gear having the speed ratio b- I ; as explained in Sect. 1.3, this would increase the effective inertia of the load, as seen from the motor side, by b- 2 • 5.3.3 Combined Armature and Field Control
From the preceding paragraph it is seen that armature and field-control of De motors each have their merits but that their normal applications exclude each other. Below base speed Wo the main :fl.ux is best kept at nominal value PeO while the speed is varied through the armature voltage U a . This is called the base speed- or armature control range. When the nominal armature voltage bl a O is reached, a further speed increase is only possible by lowering the l\laill flll X, thus creating separate field-control regions which extend the base ;) I w l ~ d r a ll g (~ iII cadl (lirectioll of rotation . Thi s is depicted in Fig. 5.7. li. i ~: fI(' I'1I I.ltal.
5.3 Steady State Characteristics with Armature and Field Control
5. Separately Excited DC Machine
(i'2.
e UaO
4>e 1
- - - __ I I'
1
e = const
I
UaO
,, "
,,
e = const (J)
,,
--UaO
roo I
_
' L ____/ . : "__ _
"
1
,
Field weakening ' I Armature range ----~,~~I· contrai range ~
- -rr-.- -
W
Pe PeO -
(5.16)
-=--<1'
4>eo
,/
I ,
I / Field ~~/~---
'I'
t----~
Wo
63
,
this causes the desired variation of the flux in the field weakening range (see also Sect . 7.3). The operating regions with armature and field control in the torque/speed plane are displayed in Fig. 5.8. The curves ia =const. are represented by parallel straight lines in the armature control regimej in the field weakening region, the torque at a given armature current is reduced, because of mM ~ Peia. For the sarne reason, the curves for U a =const. become steeper when the flux is reduced. ln the upper right hand comer a typical commutation limit is indicated belonging to a reduced armature current limit when the field is weakened.
weakening
range
-'l"eO
Fif.';. 5.7. Control ranges of separately excited DC motor in steady state
(J)
Field contrai u a == u aO 4>e< 4>eo Armature control
wil.l! the sarne polarity of the armature volt age by reversing the field (dashed cllrves); at the sarne time the torque would be reversed so that the motor olwrat.es in the third quadrant. However, this method of "field reversal" which wa~; frequently used in the past for cost reasons is rarely employed today :I,S i l. calls for the de-excitation and subsequent re-excitation of the motor; I.hie; n~qlli.r es time in view of the large magnetic energy stored in the field .. i)'nlil.. rI' lligh forcing volt age is applied to the field winding to accelerate the !III X n ~ v(~rsa.l, considerable voltages are also induced in the armature winding 1'11' 1;1.llf~( ~rillg the commutator. AIso, the armature current must be blocked dllrill!', I.he time of flux reversal causing an interruption of the torque. With 1.11/',1')' !IIotOl'S, the transition time with zero torque may be up to Is which i:: IIII!. acccptable for a fast reversing drive. It is for these reasons that drives wlticl! are cxpected to operate with both directions oftorque and speed ("four qllOldr;wt drive"), are today usually equipped with an armature power supply pl'Oviding both polarities of volt age and current ("four-quadrant DC power :,lIpply") and capable of operation in all quadrants of the ualia - plane, Sect. ~) . I; l.lIi s ha.s the advantage that the polarity of the main flux can remain Illldl
,
• 2- Troo
I I
,
/
:::r-
I .ji I
§I
~
0., I
-uaO < ua < uaO I
I II I __t1J I
__
II I 0_t1J 1
4>e= 4>eO
-2'71:
07 -1
Field contrai Ua == -UaO 4>e< 4>eO
Commutation limit
I
I
I
Ua = UaO
§I
I
0- I
II_I ....:u· • .'t1J' 2 ~
._ ~
,
\ili \
\
\
.
\
1-2
'
'
/
I
,
mM
ua- O mn Ua = -uaO
I
,I
Fig. 5.8. Operating regions of separately excited DC motor in torquejspeed plane
The figure clearly shows that the DC motor represents a linear control plant only in the armature control regionj beyond base speed, there are con siderable non-linearities even in steady state condition. Simultaneous field- and armature-control is sometimes proposed as a means of reducing motor losses through diminished copper losses in the field windings and iron losses in the armature; of course, this is only applicable at very light load when the increased conduction los ses in the armature are not masking this effect. AIso, this may cause an objectionable delay in building up motor torque, should an unexpected load surge occur, because the flux would first have to be raised to full value. Hence, this practice can only be a lIIurr;inal option in few applications, such as battery supplied electric car dri v(~s, w II\~ rI ' (~I\( ~J',.;y conservation is of overriding importance and where the dyll;Ullil' drawll1l.('h :: ;I.r<: 1I0t. ohjcdionahko
5. Separately Excited DC Machine
(iI!
5.4 Dynamic Behaviour of DC Motor with Constant Flux
;:).4 Dynamic Behaviour of DC Motor with Constant Flux
I ()=F()U () a S
III order to have a clearer view of the transient behaviour of a De machine the IJlock diagram in Fig. 5.4 is simplified by assuming the armature voltage to be :ln impressed and independently variable input quantity, the load torque mL ;\.s an independent disturbance and the flux Pe a suitably chosen parameter. This results in Fig. 5.9; the angle of rotation is again omitted for the time Iwing, as it does not affect the operation of the motor.
a S
3 S
+
F()M ()= TmsUa(s)+bML(s) 1 S LS T m S (Ta S + 1) + b2
65
(5.18)
Ua(s) is the independent input or actuating variable, ML(S) is a disturbance, also assumed to be impressed. il(s) and Ia(s) stand for output variables, which appear in the frequency domain as linear combinations of the input variables Ua(s) and Mds). All partial transfer functions of the multivariable plant have the sarne denominator, N(s) = T m S (Ta S
+ 1) + b2 =
T m Ta (s - sd (s - S2) ,
(5.19)
mL
mOn
Ua
which is identical with the characteristic polynomial of the pertinent linear differential equation. The zeros of N(s) are the eigenvalues of the system,
li)
LlaQ
LlaO
Io'i~.
2~a [-1 ± )1- (2b)2Ta/Tm]
S1 ,2 =
e
the natural frequency is 5.9. Linear block diagram of DC motor
W
Assuming the main flux Pe to be a constant parameter, a linear sys I.em rcsults which may be described by transfer functions; with the Laplace I.r;I,IIs[orms
!
e=
b
~= .JTaTm '
/'(:r(/.))
=
D = - Re(sl) = ~
ISll
x(t)
dt
e-Bt
= X (s),
(5.21)
and the damping ratio
00
2b
rr;;:
(5.22)
VTa .
Field weakening (b < 1) causes the motor to respond more sluggishly to control or disturbance inputs as is manifested by a lower natural frequency
S=r7+jw,
o
,)1' !.II<' IIo11dimensional variables
I,
'1/,,, )
(
-
u"o
{, ( ,1. (1 )
= Ua(S) ,
= Ia(s) ,
1',,0 ;\.11.1
l.h( ~
L(:) L
r-2~a
= il(s) , b max = 1
(:~) = ML(S)
jw
o
1s, 1 1
À I I I
llonnalised field parameter
---7<-~--~---X
bmin
(/Jco
[,II<'
(5.20)
I
• O'
b min
I I
l'oll()wi])l~ (~'lllil, l.iolls aH'
t
olJlaill<'d
I
J1.( ,'/ )
1"1 (.~) II,, (.~ ) I I",{~) tvI/,H
/1/1"( "'1
('1:,,'/ I. I) M,.( s)
'/'", II ('/ : ,
II
blll/l x = 1 X S,0
I I) I ,,~ (!) I '1)
Pi ,r. r. . IO.
1()1)1.
1,)(',,::
"r
1)(:
IIlOl.or
for dilr"I'I'''1. li,'ld facl ,or>< " .
5.4 Dynamic Behaviour of DC Motor with Constant Flux
:.. Separately Excited DC Machine
lili
lilld illcreased damping ratio. Fig. 5.10 shows the root lo cus, i.e. the lo cus of 1.1... eigenvalues in the complex plane, for different values of the parameter b. lte;d ('igenvalues, corresponding to asymptotic transients (D > 1) exist for 'l~"
/'~~-:-1-~0--=:;;---i\7a 2' - ,2 I"
000
/
:
//_.....
--- ""-_______________
./'" f----.
> (2b)2 Ta .
1.11 Fig. 5.10 it is assumed that the motor exhibits below base speed (h",,,,,, = 1) a damped oscillatory response. This may in practice be the case
ir
00
,, 1 0=1 b =."'F,ff;'; ,2
, \b= l' 1
i.J\(~
inertia of the drive is deliberately kept smalI as with servo drives or 1'011 iIIg-mill reversing drives but in most cases the eigenvalues of a drive are ITal Ilccause of the added inertia of the load. III Fig. 5.11 the step responses of the motor speed and the armature cur )('111. are depicted when the motor, initially at standstill, is started without I()ad (mL = O) at t = O by switching on the armature voltage U a = 0.1 Uao. i\{1.(~r attaining its steady state speed, load is applied at h in the form of con s!.allt torque, mL = 0.02 mo, corresponding approximately to 0.2 of nominal I.orq Il e. The transients in Fig. 5.11 have been computed for three different values (II" II, which result in the damping ratios D = 1/"fi, 1, "fi. It is seen that field w(~akcning makes the motor much more sensitive to loadj at higher torque 1.1t(~ speed under load would even drop below the value at full flux (b = 1), ;(.s indicat.ed in Fig. 5.6 a. However, the initial slope of the speed at tI is UI<' saltJ( ! in alI cases being solely dependent on the nominal mechanical time CIlII :il.allt.
li
d'
Tm
,
( 1.<.1) I Wo
= _ mL
~ I- f(b) .
Tili:: i:; illlllwdiately understood from inspection of Fig. 5.9, as the current I" , i.('. I.h( , lllotor torque, is delayed by the armature inductance La. TI\(' sl.( ~ ady-state value of the armature current for no-Ioad condition (ne I',i<-dill/'; I"rictioll and windage) is zerOj this is apparent from the factor s in I. IH · 1IllIllcra!.or of the transfer function F3, Eq. (5.18), which indicates a dif r( ·J( ·III.i;d.illg dfect due to the cancellation of the applied armature voltage U a 11'y LlI(, illdu ct!d voltage e at no load. II" Lhe loatl torque contains a component rising with speed, Fig. 5.12 a, ·III. /, '111'11
k
'T nLI
-
W
Wo
hloek diagram in Fig. 5,12 b results; the integra!'o}' JlOW possesses pro p()rl.ional ne(~at.ive feedback, assuming the prop( ~ r!. ks ()I" it lag d emento The diJr(·J'('llliat.ing efI'ect ill Eq, (5 ,] 8) disapJl( ,a.rH ill I.lJl' pr(';'; ('lI('.(' of loa
I"
'"
j
II
'"
II
iI
~
l . I, I,·', ~ l rI -' ~ I', , I ('t' 'IJ'J, ~ ~ -., ''~ I h
-
1
i2 -
........
, \)a '
---,
_____
n
I
ti
..!.L iaO ,",
; " :/'\ '. 0,05
\ ,
\
.,.,~ ...
b
"
mL -
,
" '..
". ......
-
°
02
---_._
.....
......:.:----------
mo~' .~~~------------~-p-~
",
-"- ...... .... -
11/ 1 ' ti
b
with k L > _b 2 and kL > - T m/Ta required for stability, The damping ratio DI may be referred to the damping ratio D at no load
mu
00
Wo
arctan kL
!!2L .!!L
uao
!!2L mo
Lhe
(I" ( II)
0-
- . --.----
"
I
a
(5,23)
- +'L -
rno
I
Fig. 5.11. Starting and load transients of DC motor for different field factor b ~ 1
mo T m
t,
'. ....... _
(1
' I"
'I
0,1
67
U) ~ I I)
e
uao
a F il!!; - r.,\:l_ 1)( :
b 1I>tIi.ol'
wit.ll speed- dcpendcnt load
mo
w
5. Separately Excited De Machine
fiK
n1
= D
1 ~ 1+(W)2b
r:-:-;;;:
(5.25)
6. De Motor with Series Field Winding
yl+V drives are subject to disturbances by periodic components in the supply ,11.a.ge or the load torque, causing continuous oscillations of armature current alld speed around their steady state valuesj the transfer function permits a :: i IIlple computation. When a torque consists of a constant and a sinusoidally varying component ,'-)O IIIC V(
'ln r-
mLl - - = --
mo
'11/, 0
~ +mo
coswmt
('ae h variable of the linear drive will in steady state contain a periodic com plllll'llt of the sarne frequency that may be computed by linear superposition. Wil.h the complex expression 7nL
mo wc
COSW m
t
~ JWm . . = --[e t + e-JWm t] 2mo
lilld for instance from Eq. (5.18) the peak value of the alternating current
COlllpOllent
/h " 'i" ,, ' -
(kL = O)
IT
m Ta (j w m )2
b
+ Tmj W m + b2
I ..1mL mo
(5.26)
;; ;"Iil;(r! y 1./11 ' plta.se of the oscillatory current may be determined. If the fre
lwriodic disturbance lies near an eigenvalue of a poorly damped condition could exist causing large amplitudes of the oscil l:d,II I" Y ('III'I"C ' II(. all(l speed deviations. '1'11 1' dYII
'111( ' '' (''y .. r ;\.
"I I v(·)
ii
rc 'sollallt
This motor differs from the one previously discussed only by the design and the connection of the field winding which, according to Fig. 6.1, carries now part or all of the armature current. Ral is the armature resistance, possibly increased by an external resistor, Rp is an adjustable shunt resistor for field weakening. Because of their operating characteristics, series-wound motors of larger rating are restricted to traction drives. ln urban transportation, such as trol ley buses or subways, motors up to about 200 kW are used but for main line locomotives motors of 1000 kW are common. ln principie, series wound motors can be fed with either direct or alternating current but the construc tional design would be quite differentj for example, a laminated stator and special provisions with regard to commutation are required for the AC mo tor. When AC traction was developed at the begin of the century, the current induced by transformer action in the armature coil temporarily shorted by the brushes during commutation, Fig. 5.2, caused problems, which were then solved by choosing a lower supply frequencyj this is the reason for the 16 2 la Hz power supply used on central European railways until today. Since com pact and rugged power electronic components have become available, main line locomotives with AC supply were often equipped with rectifier- fed DC motors. ln future, all these schemes are likely to be supplanted by AC motors fed from variable frequency inverters.
ia-ie
mM,W,E
ie Re
cI>e
Ua
Rn-R,11
Fi)·:. O. I. 1)( ' "" .<. .... wi<.h ,,, 'ri "H fi..!d Willdillf(
'lI!
6.1 Block Diagram of a Series-wound Motor
I)C Motor with Series Field Winding
(j,
~ ; lIlitll
ltigh-speed AC series motors, so called universal motors, have a wid(' fidd of application in household appliances and power tools. 'l'lw typical operating characteristics of a series-wound motor could also 1)(' rcalised with a suitably controlled separately excited motor; this can even 1)(, prcferable in view of the added flexibility, for example when a traction 1II0l,or is supposed to serve as a regenerative brake.
Tal
d dt
(La
l
Ral
T
1
e
!i (~) dt
1'(jllaLions describing the dynamics of the series-wound motor are similar c" LlI()SC derived in Chapo 5. The field and the armature windings are orthog (III:'" Lo (~ach other and hence are magnetically decoupled. Due to the large air!·,i l.p iII the armature axis, saturation is neglected and the magnetic circuit i" (1'ladrature direction is assumed to be linear. R' a ta
-
R p ('ta
,)
- te
-
e,
(6.1)
" dPe ,TV,, - = - R' e te
+ Rp (.ta -
dt
e = c
=
lnM
.I
, ), te
(6.2)
T
c
Re Ral Re
+
ml
d dt
(w) = WI
T
ie
Ra Ral
+ Re
-
_
La Ral
+ Re '
aI -
i al
~+~ i al
Ral
ial
(6.8) '
mL mi
Tml
=
JWl mi
(6.9)
.
:t
(Ne
+ L a ia) = Ua - e -
(Ra
+ R e ) ia
(6.10)
.
(6.5)
- = 'f/I,M - mL . dI.
1) is defined at nominal armature volt age I. wit.JIIIIIL lidd w( ~akcning (R p --+ 00) and with no external armature resis I
mL
1/ "
1",Jd "
mI
..!!iL Ual
(' ,
C '/'",
"' II I
'l ' I
'/ II, I
I:
(
w, =
= (1 -
J.L
'T}l Ual
Ual
'T}t)
iao
e
uao
-;:;----+ Re ai
iPt: 1 i al .
(', / 11" I i~ tlte electrical efficiency at the nominal operating point (not for iron- and friction losses). These quantities are used for nor III:di s llq'; L1l( ~ Eqs. (6.1- 6.5).
'II
1 J( " '''11I1LiJl !~
'I." 'f'ul
Pp. )
Jo: (
(6.7)
ia - i e
+ Re
Normalisation, as before, yields
Ilw
ia i al
A graphical representation of these normalised equations is seen in Fig. 6.2; the continuous state variables appear again at the outputs of integrators. The effect of the parallel resistor R p in coupling the two magnetic circuits is clearly recognisable. Without field weakening, i.e. Rp --+ 00, ie = ia follows; armature and field winding are then forming part of a common electrical circuit, hence
(6.3) (6.4)
,
+Re
w)
'T}l
ia -
N e
'1'1",
di = U a -
1 ( Ua = 1 - 'T}l Ual Rp
H.l Dlock Diagram of a Series-wound Motor
.Ia dia
)
71
ia I cPe
(6.6)
CPeO
I:: !.I", lIul.glldisation C1lfve in the longitudinal axis oi' the machine. As i I, r('s llll" t.ltrc~(' (liuH'Ils ionless (eXC!lpt Liul(') fir s l. onkr s Latc cqllatio lls r,,1' LIli' 1.111'<'1' elll'I'!';.\' ~; L(lral~c va riahlcs 'i ,,, 'P" , I." :1.11' o!.l.ailu "I:· 1"11-\ . n.~ . 111", '1, " " 'I'/ IUII
(Ir n 1Wl'i,'!j- W
1)( ~
IIllIl,"f
.,
DC Motor with Series Field Winding
fi.
d (Pe
"
dt
.fel
Pel
La i a1 ia) i a1
+ N e Pel
= 1-
1
6.2 Steady State Characteristics
( "h
Ua U al
-
7]1
Pe Pel
w)
6.2 Steady State Characteristics
Wl
Ra + Re ia Ral + Re i a1 . 1kc,lIlse of the larger airgap in the quadrature axis, La i al
(/ = N e Pe1
_ !lial - !lie1
«
73
The steady state behaviour of the motor is again found by putting the right hand sides of Eqs. (6.8 - 6.9) equal zerOj with Eq. (6.12), a linear magnetic circuit is assumed in the longitudinal direction of the machine. Elimination of ie/ial' results in
1.
-U
a
U al
=
[w 7]1 b -
Wl
+ (1
-
7]1)
r
]
.ia Zal
(6.13)
WiLh Eq. (6.6), noting ie = ia, Eq. (6.10) is transformed into
where
w)
( -Pe )] = -1- ( -U a - 7]1 -Pe P el Wl Pel 1 - 7]1 U al Ra + Re ia (6.11) fl al + Re i a1 The associated block diagram is drawn in Fig. 6.3 for q « 1. A character isLic feature of the series-wound motor is the dependence of the flux on the ;U'JIlature current, which causes it to vary with the load. Since most of the ali 11 ature voltage is balanced by the induced voltage, U a ~ e = cP e W, high ".pccd is to be expected at light loadj this is reflected by the steady state CII rvcs discussed next. To gain a better insight, saturation of the iron may be ignored which in I"ig,~. 6.2 and 6.3 is achieved by omitting the nonlinear functional block, Eq. T e1 -d [Pe dt Pel
+ q fe
(lU.), ,i "_. = "." I
f (~) Pel
Pe
(6.12)
Pe1
t\ I. h il ';lI cunent, when saturation cannot be neglected, the motor character d .ícs
b= ie = ~ ia Re + Rp is a factor characterising field weakening and
Ra + bRe R a1 + Re
r=-----...:.
,-
- -----
,I
ia i a1 =
..!..!.L.
mL
m..
1(1/1
11" ,
()
11. 11
','IH . lI . lI . l!l,,, '" dlil !", li 11/1 " I rll·r h·./l
IVlllwd
II< :
Il u d ," , '
-V IU"," ', II" L
W
R a1
+ Re
(6.15)
(lmL
Vb -:;;; .
(6.16)
Eliminating ia/ial from Eqs. (6.13), (6.16) finally yields the torque/speed curves of the unsaturated series-wound motor ~ tLal
_
Wl -
ia1
R+~ a Re+Rp
is the ratio of the effective armature circuit resistance to the nominal re sistance without parallel and series resistors. Eq. (6.13) indicates that the armature circuit of an unsaturated series-wound motor appears like a resis tive circuit, whose effective resistance depends on speedj this is of course due to the induced armature volt age e which is proportional to armature current and speed. Likewise we find from Eqs. (6.5,6.9) with mM = mL
W
J.L
(6.14)
7]1
1-
7]1
r
Jb!!!:k - ----:;;;- b ' m,
(6.17)
where the normalised armature voltage U a /U a l may be continuously vari able whereas field weakening (b) and armature resistance (r) are subject to discontinuous control actionsj the load torque mL/ml represents again a dis turbance. A set of these curves is shown in Fig. 6.4, calculated for 7]1 = 0.9. As ex pected, the torque-speed curves are very steep at light load. This is desirable for vehiclp. rlrives, because the motor can be operated over a wide speed range anel sharill('; Lhe load on multi-motor drives is no problem. The similarity of t.ltc CII)'V(' S L() I.II( ~
6. DC Motor with Series Field Winding
7'1
6.2 Steady State Characteristics
'l'lte speed rise at no-Ioad could of course be a source of danger in other than I.raction applications. The armature current/torque curves, also depicted in Fig. 6.4, show a silllilar increase at field weakening as was observed with the separately excited IlIotOr. ln view of the higher copper losses in the armature and the more nitical commutation at reduced field, the amount of field weakening must be I(~stricted here too. On street-cars, trolley buses and subways, where normally a constant di rcd. voltage is used for supply, as well as with battery operated vehicles, :iwitched armature circuit resistors may be employed for controI. ln recent )'('<\rs electronic choppers have been introduced which permit continuous vari ;\!.ion of the mean armature voltage. At higher speed the field may be weak ('lIcd by parallel resistors. ln addition, if several drive motors are present, they (";111 be connected in series or parallel, thus providing a simple way of changing I.he voltage in a few steps. With city trains or main line locomotives having an /\ C catenary the armature voltage can be varied in fine steps by transformer I.a!> changer or continuously by power electronic devices; rectification may or 111;\'y not be included. To reverse the torque, which is proportional to Pe ia, a reversal of the field willding is required; the machine can so be operated in the motoring, i.e. the lirsl; anel third quadrants of the torque/speed plane. When operating in the ti('coml and fourth, i.e. the generating quadrants, an unstable condition exists wliic" is manifested by a horizontal asymptote of the torque/speed curve at n ("lil.i(";\1 speed, Eq. (6.17), Lv /" ( 1. '1
III
l-r/l~
This constitutes a state of self-excitation, where the effective armature resis tance, contained in the brackets in Eq. (6.13) becomes zero which means that the armature, being driven in the second or fourth quadrant by an active load torque, is supplying the los ses of the armature circuito The result is a steep increase of the braking torque produced by the machine. The critical speed Wc is a function of the total armature circuit resistance and field weakening but is independent of the applied armature volt age. The effect of self-excitation is seen in Fig. 6.5 showing a calculated tran sient with an unsaturated machine. At first, the motor is accelerated in the first quadrant at no load with a small value of the armature volt age U a until time t l , when a constant load torque mL is applied which decelerates and eventually reverses the motor into the fourth quadrant. When the critical speed Wc < O is reached, self-excitation in the form of a nonlinear transient sets in with speed oscillations and high peak torques which in practice could endanger the motor shaft. Since this effect does not depend on the supply voltage U a , it can be utilised as a supply-independent electric brake in vehi eles; this is described in Fig. 6.6, [B2] . mM. W
mI '
Wl
mL _ mI
!!!.b...(t) m1
\\
I'
I
V'
'>
Acceleration
at no load
(6.18)
-~ b
Wc
f4\ ia -.::..;/_ia1
Wl
- !b- ! - !r -R a
UUal
- RI
-
ie ia
(J)
®
® 8)
Load applied
mL mI
=0,25 =const
Nominal operation
Fig. 6.5. Transient of self-excitation of a series-wound motor 0,5
1 1
I
4 1
I
0,5
p
- p; =const 1111
"" t< ' liA .
75
:-1 1,' ·II .~ y .'I l.Id." .. l. ar l"·C," ill l,i" r! I,r 11 11' \1' 1" 11
\V""""'
1)1 :
,,,,,1.,,,'
It is assumed that the field winding has been reversed and the circuit is elosed through a braking resistor. The motor is running with W > O, so that braking calls for operation in the second quadrant of the torque/speed plane; it is controlled by the effective armature circuit resistance, Eq. 6.15. A detailed analysis of the braking transient requires the inclusion of iron saturation, but the principIe may be deduced qualitatively from the curves in Fig. G.a h, WIICl'C e(ia) is the induced armature voltage at a given overcriticaI :·qH'd nlld U. '':" i:; t.IlC voltage across the total braking resistor including the 1'(·:1i: ;1.1\1I("(· 11'." 1 ,"I'!.lll' arlllat\l),(~ itsdf. III Lhe initinl pha..se, the current is elriven II,Y 1.1", 1" '11 ' fi 111'11' ,. 1'1111.11.1',(' ('II, wil,ltolll. ,'x·I.('J'lIal V()It. nl ~(' lwillf~ appli(~d . Tlw
7(;
6. DC Motor with Sedes Field Winding
e, Ria
e(ia , (O) (O
R
= const
(O
e-R,'a--N d
-- el a)
b)
dt
ia
Fig. 6.6. Short circuit electric brake with series-wound DC motor. (II) Circuitj (b) Characteristics
ral.( ~
of rise and the eventual stable operating point Sare determined by the
(I iIkrcntial equation dPe N .e -
dt
= e ('za, W ) -
7. ContraI of a SeparateIy Excited DC Machine
R'za,
;11':aill llcglecting the self-inductance of the armature compared with that of I.h(' lidd winding. ln the steady-state condition the machine then operates :t.; :I. 1': ( ' II( ~ rator, converting mechanical power to heat in the predominantly ,' xl.('1I1;d hraking resistor R. (>II Il laill lilH~ locomotives, the drive motors are often reconnected as sep dl .d.('l y (· xcil.c( l generators during braking operation, in order to have more d i I(., I. (o/llrol. The auxiliary field power may be taken from a battery, if :: lIl'pl ,\' illdq ) ( ~ lId e nt braking is specified. 1\11 I:x: l.lllplc of controlling a DC series motor is discussed in Sect. 9.2.
7.1 Introduction ln Chapo 5 the steady state and dynamic behaviour of a separately excited DC machine with adjustable armature and field voltage has been explained; this discussion is now extended by considering the machine as part of a feedback control system. The reason for this is that in practice the choice of a DC drive is normally motivated by the possibility of operating over a wide range of the torque/ speed plane with low losses and matching the behaviour of the motor to the need s of the mechanicalload. To achieve the desired operating char acteristics in the presence of supply- and load-disturbances, feedback control is usually necessary. Another reason why DC drives are normally contained in feedback loops is that the armature of a larger motor represents a very small impedance which - when supplied with nominal voltage - would result in an excessive current of up to 10 times nominal value. Under normal oper ating conditions this is prevented by the induced armature voltage e, which cancels most of the applied voltage U a so that only the difference is driving the armature current ia; these two quantities are performing the actual elec tromechanical energy conversion. ln transient operation, for example while accelerating or braking, there is always the danger of excessive current due to the rapidly changing armature voltage or sp eed; the same is true with a steady state overload on the motor. It is therefore important to provide a fast current or torque limit to protect the motor, the power supply and the load; this is b est realised by feedback control establishing an effective safeguard against electrical or mechanical stresses. At the same time an unequivocal criterion is gained for distinguishing tolerable temporary overcurrents from currents caused by a functional fault which calI for immediate clearance by a breaker or by fuses. ln most cases the user of a controlIed drive wants to b e able to select a reference speed wRef which the motor should maintain as -long as it is not overloaded. When overload occurs, the motor should produce maximum torque ill th c pr e~ crib e d direction; twice nominal torque is often specified as a shorl. Lcrlll lillliL This results in stea.dy st at,(, eharacteristics of th e tyP( ~ Slt O WII iII I"il ~ .
7. 1.
111 1l. 1l 'y I\. jljll i.-: tI.ioll S, sHch as (:!<: viI,l.()r H, 1.111':-:(' i'l jl('c ilicHI.i o IlH ar(' ill1mJ' II('i" 111. I",\ ·H.IiI II· 1.11"1'<' 11 1'(' "I)I'I·a l. iI !J ( ,·ollel il.ioll il Wh" 11 I.h,' 11.,...,·I'·I'il.l. i"lI (II ' 1.11<' 1,'01'
7. Control of a Separately Excited DC Machine
'7il
7.2 Cascade Control of DC Motor in the Armature Control Region Supply
Torque limit
ú)
cRef roRef
normal operation
79
aRef ~
ú)Ref
Ye
Rotating exciter or converter
ie
!Ue
I IRe
~
C>
-mn
mn
2m n
m
liamaxll
~
c::
C> ()
- -Ú)2 - , - - - - - - - - - - 11 1
"'i~.
c,rol
lia
lUa
Rotating generator ar converter
I lU a
7.1. Steady-state torquejspeed curves of a controlled drive with torque limit
angular position must be controlIed according to prescribed references. AIso l11achine tools or robots, the feed drives in the different axes must accu ral.dy follow prescribed angular commands as functions of time, in order to 1I10V(' the tool or workpiece along a desired spatial trajectory. Ali these requirements can be fulfilIed with the general scheme shown in Fig. 7.2 where the armature and field of the motor are fed from separate con I,mllaillc power supplies. Rotating generators in the form of the welI known Wal'd Leonard scheme have been the common choice in the past but, after l, ri,;I" 111(.('rllldes by magnetic amplifiers and mercury-arc converters, power ,'I,~,·(.r,,"ic COllvcrters employing solid state semiconductor switching elements (\';lli"II:: I.ypes o[ thyristors or power transistors) are now the standard solu
()Il
I. i" II,
W iI, h a "r()1If quadrant" armature supply allowing both polarities of volt
1\.,"1 Cllrl'(~lIt, the power flow to the motor can be reversed, with the III III 'I,ill" "perating in alI four quadrants of the torquejspeed plane. This is 1.111' (':1."', ir a rotating DC generator driven by a line-fed AC motor serves as n, P"WI'I' :-iIlPply; the sarne can be achieved with switched converters. With 1I11I/',II('l,i.: alllplifiers, containing only diodes instead of switches, regeneration WIL; : 11,,1. possible. Of course, there is always the option of electrical braking I',Y "laóllg resistors in the armature circuito 'I'he :-;tcady state power required for the field winding is only a few per ,','11(. of I.ltat for the armature; a "single quadrant" rectifier, controlIed or 1II1'·oll'l.rolkd is adeqllate for feeding the field winding. li"i/',. '1.2 contains a block "control equipment" which is controlIed by ref "I"~I((," ~;i/';Ilal:::; én,'!, wne!, CXRe!, 'ia mal" as welI as feedback signals é, w, a, I.,,, a~ i 11,:,'']('<1 for sp" cifie control taslcs. The sp eeu signal is normalIy obtailled f (' OIlI II, Hll lHll pCl'lI mllcll t 11l
Fig. 7.2. General schematic of DC drive control
each angular incremento The frequency of the pulse train is then proportional to speed and counting the individual pulses provides a measure for the angu lar position change; the sign of the rotation is determined from the sequence of pulses generated by a dual track sensor with the two signals being 90° out of phase, Chapo 15.2. The controlIers transform these input signals into the actuating variables Ya, Ye which serve as control inputs for the two adjustable power supplies. ln the folIowing paragraphs some proven control procedures for DC drives are discussed, making no specific reference to the type of power supply used, since this topic is dealt with in later sections.
7.2 Cascade ControI of DC Motor in the Armature ControI Region Let us first look at the case when the motor operates below base speed, -Wo :::; w :::; Wo. As explained in Chapo 5, the main flux is kept at the nominal value PeO to have the lowest possible armature current for a given torque and to guard against unexpected load surges. ln Fig. 7.3 the schematic diagram of the armature circuit is drawn; e a is the internal volt age source of the power supply. This synthetic voltage which is controlled by y" differs from the armature volt age U a due to the internal impcd all cc / li , / ,; 01' Lhe s1lpply circuit.; this m1lst. be taken into account. when llorJlIalisill/'; I.h(' 11I()I."r p;l.rnlJld.(~rH,
7. Contrai of a Separately Excited DC Machine
I)()
7.2 Cascade Contrai of DC Motor in the Armature Contrai Region
1----------- I I
-1:I
R·I
L·I
I
c:=Jt--I_---
I~
Ya
I
~el
ua
a e
J
La
ia Ra
00----
~-----
______ I
Power supp/y
(actuator)
F ig. 7.3. DC motor with armature power supply
La +Li
eaO
i"o = Ra
+ Ri'
Ta
= Ra +Ri
Witlt the results of 8ect. 5.3 and b = I, the block diagram of the motor a SSlllllCS
the form shown in Fig. 7.4 a, where the controllable volt age source
('" is represented by a first order lag element having the voltage gaín G as and Lh( '
81
The armature time constant Ta usually has values between 10 ms and 100 ms; it is determined by the impedance of the complete armature circuit including possibly a smoothing choke that may be needed for reducing the current ripple caused by switching convertersj with small servo motors hav ing an iron-free rotor the armature time constant could be less than 1 ms. The mechanical time constant T mn comprises the total inertia of the drive (assuming rigid couplíng) j therefore it covers a wide range from a few mil liseconds (servo motors and reversing rolling miU dri ves) to several seconds (paper míll drives and mine hoists). There is general agreement that the most effective control scheme for drives is a cascaded or nested structure with a fast inner controlloop, límiting torque or, on a De motor, armature current, to which an outer speed control loop is superimposed [K24, L79, 885]. This multi-Ioop system proves very flexible when the control plant possesses a chain like structurej for example, in order to control position, it can be extended by a position loop superimposed on the existing speed control loop. AIso, if acceleration is of importance, a corresponding inner controlloop may be added.
tiUle constant Tas, which strongly depends on the type of power supply.
Whell a rotating generator is used, Tas corresponds to its field time constant
;uld may be in the range of 0.1 to 1.0 s, whereas the corresponding value WOJdd \)e 1 to 5 ms, if the residual delay of a líne-commutated converter is 1. (1 III: Jll()(klled. Naturally this difference greatly affects the dynamics of the I ' , 111I,rol systern; it must be taken into account, when designing the controllers. I"i,... '(, /1 b i s ;LIl equivalent structure, redrawn for noninteracting access to the ÓI,rlll i d ,IIIT
c urrc llt .
Actu ator
/'(Iwnr :al/'/'/Y VII
Supp/y
mL
ia
mo
mM
OJ
~I;;;=mo
E
ú)
~I _ _ _ _ _UaO
n
Fig. 7.5. Cascade contrai structure of drive including position-, speed-, and current contrai with feed-forward from a reference model m I..
mo
I
;a mM ~t;;;;= mo
(I)
The idea of cascade control is exemplífied by Fig. 7.5, where a lumped inertia drive is cquipped with most of the controlloops mentionedj it is seen t,h at the SCqUCll CC: torque-acceleration-speed-position is a natural one as it cO llforIJlS to Lltr. strn d nre of the plaJlt. Tll( ~ torque (or annature current.) loop, f'ur;llIilll ( I,!il; illlW)'Illost (:oul.ro[ fllllcl.io ll l\my lw regankd as appro Ki Ill a l.r'ly nl 'l d, i ll ,~ :UI illl[)J '( ': ;:a'
V,,
I, I" i ~( •
[k ~,.1, 1''llli 'l! d l"nl l1 J1l f' ct t 1)(:
1111
ai.II I'
rl l,1.
"..
1I 111 1lI 11 11 1I 1'" \'\11' 1
II l'pl Y, 'I',
'/' , II
('(lIll.lolln 11:1:: 1.(, <1" 11.1 pri lll n rily w'il.lt 1,111'
1.111' pow\'r li lll'l'l ,v Il.Ild
K:l
7. Control of a Separately Excited DC Machine
Lhc armature, cancelling the effect of the induced armature voltage; thus it provides the fastest available control action. By limiting the reference mM ReJ (or 'ia ReJ), the inner controlloop also assumes a protective function. An ac ("d(~ration controlloop which is not shown in Fig. 7.5 could be desirable with I! iglt dynamic performance drives, such as servo or elevator drives; its main I"l('pose would be to counteract the effects of load torque mL as well as cl!;lllges of inertia, by generating - within the admissible range - a suitable ("llrn~nt reference ia ReJ. The hierarchical structure is then continued with ;1, spccd controller which controls the closed inner loop in conjunction with LI[(' subsequent integrator; the same idea is once more extended when going Lo ;l position control loop with the position controller generating the speed ,'('I(~r(,Ilce.
()f course, this multi-Ioop control structure can only function under the ;I,s:mmption that the bandwidth of the control increases towards the inner loops with the current loop being the fastest and the position loop the slowest; ("\('arly, the position controller can only perform well if the speed loop is (jllickly executing the commands it is given, etc. Following this reasoning, the dcsign of the controllers proceeds in the direction from the lower to the higher ('()lIl.rol leveis, at every stage approximating by a simple model whatever has I)(~(~n a.chieved before. Obviously, the design of the control system is greatly :lÍlllplincd by such an iterative procedure, where only a part of the complete pLtlIl, is dcalt with at a time. The fact that each feedback variable can be lirllil.(~d by damping the pertinent reference is a major advantage of cascade 1'11111,]'01; a.l~o) ficld work is considerably simplified, since the controlloops can I,,' plll. ird,o op<,ration and tested one after another. II, rll:t,y IJc aq,;ued that a control system containing several inner loops is I"",Iy 1.11 IT:;pOlld slower to changes of the reference than an equivalent single 1,,"1' ("olll.rol (provided that it can be stabilised). Therefore, when dynamic 1"',lmIJI;\lIc,; is illlportant, e.g. with position controlled servo drives, the inner 1""1':: ,':lllIldd be activated in parallel by feed-forward reference signals derived 1111111 :\.11 "xt,cmal reference generator as indicated in Fig. 7.5; as long as the I('I"I('II('C I\(~llcrator is outside the feedback loops, the stability of the control :;y::L"1I1 ('clIlains unaffected. At the sarne time ali disadvantages with regard to dYll
83
7.2 Cascade Control of DC Motor in the Armature Control Region
after the cost of control amplifiers was dramatically reduced by transistor operational amplifiers and electronic integrated circuitry. When comparing the structure of the plant in Fig.7.5 with the block diagram of a DC machine in Fig. 7.4 a it is seen that the machine-internal feedback of the induced volt age is not immediately compatible with cascade control because it represents a deviation from the nested loop structure of the plant. However, by redrawing Fig. 7.4 a in the form of Fig. 7.4 b, a non-interacting structure is obtained; the fact that the load torque is now entering the plant in two places is no problem since mL is assumed to be an independent disturbance. The variables not specified in Fig. 7.4 b have no immediate physical significance. The part ofthe plant that lies between ea/uaO and ia/iao may be described by the transfer function
~(8) Ea
=
Tm
Tm Ta 8 2
(7.1)
8
+ Tm 8 + 1
corresponding to F3 (8) in Eq. (5.18) for b = 1. The differentiating effect vanishes with speed dependent load, as was seen from Eq. (5.24). The design of the current controller is explained with the help of Fig. 7.6 assuming no inherent damping by the load. The feedback signal derived from the current transducer is passed through a low-pass filter to reduce possible ripple; a filter time constant of 2 to 5 ms is usually sufficient. mL
mo -----1 Motor I Current controller
Actuator
I I I I
Tms
Tm Ta 8 2tTm 8t1
I.
'
-f'ª- = mM mo {aO
Sensor
Fig. 7.6. Design of the current controller, Pc = p.o
If a POW(T dcctronic converter is clllploy(~(! as armature supply, a pro.. port.ioualilll.(·I':ral cO!ll;rollcr (PI) i::; t\
1\1
7. Control of a Separately Excited DC Machine
7.2 Cascade Control of DC Motor in the Armature Control Region
Zener díodes
and the applied voltage rise linearly with time; with other words, generation of a ramp function with a controller containing a single integration calls for a constant control error. For the practical application it is normally of no consequence whether the current controller admits a steady state control error during acceleration, because current control represents an inner (auxiliary) function , being part of the superimposed speed loop. The main purpose of the current controller is to linearise the plant and prevent overloading the motor; at constant speed, this is accomplished by a controller with a single integration. It would in principIe be possible to use a current controller performing double integration, e.g. a dual PI-controller having the transfer function
Clamping circuit Feedback circuit
7]C4 Reference signal
~ Rs
RI
X1i
Y
Actuating
signal
1
Fb,i(S) = GC,i T l S + 1 T 2 S + 1 Tl S T2 S
wíth
X21
Rs « R4 «R3' C3 « C4
Feedback signal
y
(T I s+1)(T2 s+1) F (s)=-(s)""G c XI c T I s(T '2 s + 1) R3
,
Gc ""-, T I ", R3 C3 , T2 ""R4 C4 , T 2 "" RS C4 RI
Fig. 7.7. Circuit of analogue PID-controller employing an operational amplifier tlw controller function by an analog operational amplifier, this is a minor ildditi()\l t o the RC-feedback network of the amplifier. A typical example of /1.11 all;tl() ~ \le PID-controller is shown in Fig. 7.7. Wil.h tb ese assumptions the open loop transfer function of the current (' ()) tI ,]" () I bccomes PI
PDT
r---A--. r---A--.
, 1'11/' ,; (8) = GCi
T l S + 1 T2 S + 1 Gas T l S T~ S + 1 Tas S + 1
'-------.,,,...------' ' - v - - '
PID-controller, T~ ~ T2 Supply
,
Tms T m Ta s2 +Tms v
Armature
85
1
+ 1 Tfs + 1
(7.2)'
'''-v--' Sensor
lkcau:ic of the factor s in the numerator of the armature transfer func a jJroportional-acting cJosed loop results, lim Fods) = FodO) finite, s-+O (, V( ~ II I.h()lIgh the controll cr itseJf comprises an integral term; as a result th e c!()s(,d ('.IIITI,'I11. r ()lIl,rol loup will exhilJit a steady state erro r. The reason is III' f'(l\ u r~ (\ I.hl' ,(' .11 I. f. r(,,'dhadc which rcmaim, dfcctive also ir; t.h( ~ l,r: tll li f; )l'IIJ( ~d I,l"d!, cl l(l.,~f·'~ 1 11 "r I" I I ~ '(, II h. C l{~arly, wlll'lI :; lIpplyill!,: (·()II Hl.al1 l. 1' ( 11'1" ' \11. I.() UI(' IlH ' l l d.l\ II' 1,1' II UI : 111 (11.... ' 11. 1. "" lClfv l, 1.1 .... "",1."1' /1 )""'11, 1.1i. , illdl W I'cI VI .!!.;II',"
(7.3) '
for avoiding a steady state control error during acceleration. However this is not normally done in view of the reduced margin of stability. If maintaining a prescribed acceleration is a necessary condition, as may be the case with machine tool- or elevator-drives, it would be easier to include a separate acceleration controlloop as will be shown in Fig. 15.9. The choice of parameters for the current controller according to Eq. (7.2) foJlows the usual lines for the design of cascade control systems [K25, 38] . ln the case of a Ward Leonard scheme containing a rotating generator and with aperiodic damping of the drive (Tm > 4 Ta) the lead time constants T l , T 2 of the PID-controller could be tuned to Ta and one of the emerging lag time constants of the drive, while the rest can be dealt with by approximation on the basis of the principie of "equivalent lag" , e.g. [38]. The closed current controlloop is now inserted into the speed loop form ing the next higher levei of control, as is shown in Fig. 7.8. Provided the current loop is well damped, it is acceptable to replace its closed loop trans fer function by a proportionallag element having the gain Gc L,i and the time constant Tc L,i' Since this equivalent lag, being a first order, low frequency approximation of a much more complicated transfer function, cannot be sim plified by lead time constants, a PI-controller for speed is the appropriate choice. Hence the transfer function of the open speed loop is approximated by
FOL,w( S) ~ G ow
1 Tc,w s + 1 GCL,i T c ,w S TCL , iS + 1 T m S
(7.4)
'-v-'~
PI· Controller
l.i()lJ,
Current control
The pa rameters G c , w , Tc, w may be chosen in accordance with the "symmetri cal Opt.illIHIlI" IX 2:i,L33,38], which is a standard design procedure for transfer fllJldioll s cU Ilt.a i II i II g a double integratioll (indllding the controller). '1'1", ll lll,i ll i el" II, ln I.() c11()W, (~ t.h(~ cross-()vc'r frc'CjIl('IWY at t\w gC'ornetri c IIlca n III' 1.11<' I.w" ,'""," (','
r... ''I ""ll(' i('H, 1.0
ohl.;I.ill 111 11 Si 11111111 ph lls,' IIIi1rl ~ llI 'l/I,{ whi ch iII
7, Control of a Separately Excited DC Machine
r((\
7.2 Cascade Control of DC Motor in the Armature Control Region
Limiter
From the cross-over condition mL
mo
.!!!1..
(7.7)
!FOL,w(j wd)1 = 1
mo
the gain of the speed controller is obtained with Eqs.(7.5) ,(7.6)
(Ollol
("0
1 Tm G c,w = aGcL ,i T CL,i Speed control/er
"'i ~ .
Current controlloop (Approximation)
ctJe = ctJeO
7.8. Speed control with approximate representation of current controlloop
" úJd
- 1
/" 1I
(J)
FoOw)
1 I
II
a;
1±j
J (a; 1 1-
1) '
(7.9)
~ Pha se margin
b
1/
(7.10)
is the damping ratio of the oscillatory part of the response. Assuming a slightly underdamped response, D = ljV2, we have a = 1 + V2 : : ,; 2.4l. When aperiodic transients are specified, D = 1, leading to a = 3, a triple real pole at s = -Wd results. Due to the double integrating term in the open loop transfer function, the closed speed loop exhibits zero control area or, synonymously, zero velocity error [38,88]. This means that the step response is characteúsed by consider able overshoot eventhough the transients are well damped. To eliminate this effect which is caused by the lead term of the Pl-controller, a corresponding lag term
Re
I I Cross-over frequency
'PI)
D = a-I 2
jlm
' ""--2
I I
".' ~ w, [-
/
W
1
1
~ -w"
Hence
aWd
I........
(7.8)
The poles of the resulting transfer function of the closed speed control loop, i.e. the eigenvalues of the speed control loop, can be ca1culated explicitely. They are, assuming a conjugate complex pair, positioned on a circle in the s-plane with the radius Wd,
I' " I
II
87
"' ll\· 7.!.1. Bode-diagram (a) and frequency response locus (b)
1 T c ,w s + 1
n""ord i llg to "symmetrical optimum"
t lll"ll wiU result ín optimum damping of the speed loop. The name stems chc Dode-diagram that shows symmetry with respect to the cross-over Ilt'qll ( ~ ll cy, Fig. 7.9. Il f llorrnalising with
1'1"1111
'/:.)(1)
Ll1I'
(1,'2
'1 (,' I"
i ,
('ro S~H )V( ' r f'y qlll'llry
/";1
) '1'I
, i"
'I' f
' I} , i
a > 1,
(7.5)
is n '/~A' l , . ,
(7.1i)
may be inserted into the reference channel of the speed controller. ln practice this can be done by connecting a suitable capacitor Cl between a center tap of the input resistor and ground, as is indicated in Fig. 7.7. The speed controller shown in Fig. 7.8 contains a nonlinear feedback block which Jimits the current reference to the range -ia ma:c :S iRe! :S ia ma:c' This serves as a n effective protection for the power supply and the drive, pro dUeillg tll P desired steady state torque-speed characteristics shown in Fig. 7,1Ü; of (,()lIr s ( ~ , a :;)Iort overshoot of the current , for example following a load disl.llrhall(,(· , C l lI llPI. h(~ rnled out hy dalllpi,'1!; t11C eurrent. rdcrenee. If the l'It '('j,f(lIti. · cOlrl,I'"II,,!' i,: lI'a li s(·d wit.ll t!te IJI'lp /)1' nll opt'r:üiollal a'llplifi( ~ r , 1.11(:
HH
7. Control of a Separately Excited DC Machine O)
Speed control/er clamped
0)0
O)Ref
0)0
o
_TI
iam~ICD
I" fIItJX
('o
iaO
7.2 Cascade Control of DC Motor in the Armature Control Region
transients, is seen in Fig. 7.11; it restricts the occurrence of current limit to load surges, for instance when a reversing rolling mill is mechanically jammed because of a wrong screw- down setting calling for an excessive reduction of the billet thickness. On the other hand, current limit is not reached just be cause an impatient or careless operator rapidly changes the manual speed command.
dJ.L
J.L+Ta ~ ,I
"
ia Ref
iaO
'aO
0,5 "'ig. 7.10. Steady-state torque- speed characteristics of speed control loop with 1.1)"'1I1' ~
lilllit
0,4
dt
/~
ea --
//.
I I
,
UaO/
I
0,3
td / W.f/u l 1rJ1)
rl~
:I
I ... r--;~--F--
I ... úJRef II --/ ,I
' I ,
úJo
0,2
=
0, 1
° °V
0,5
___
I 'uao
_______ _
/
I, '"
cPe =cPeO
UaO
. Start at no load
0,2
ea -
,I ,I
I1 II
/
..1;\lIlpiug can be achieved by placing Zener-diodes in the feedback branch, I"il~· 7.7. Limiting the current reference removes in principIe alI restrictions "II I.he rate of change of the speed reference, which is important if it is pro dllC(~d lllanually. Whenever the speed controller, due to a rapidly changing "')(lIlllal1<1, reaches clamped condition, the speed loop is disconnected while Uu ' Cllrrent control loop remains operative with clamped current reference. fi:; S ()OI1 as the speed has caught up with the command, the speed controller I"I'I'I'I.S 1.0 linear operation and resumes controI. Ilow('ver, with larger motors under manual control, a rate-of-change lim it" ·,, is ()rt.( ~ 1l illserted in the reference channel ofthe speed controller to ensure 1.1, :11, 1.11<' (,111J'(~llt limit, which may correspond to twice nominal current, is not "'./1111'.1 1IIIIICn~ssa.rily. A block diagram ofthis device, together with typical
89
~
J.L
I'
iao
"
1,0 1,5 Load applied
~ s
· '. -iamax .-=0,2 LImlt. 'aO
"
ia Ref.J"
0,1
lo' a ,
,
úJRef
úJo
I
°
I
°
I
II
'" ._ 0,5\1
1,0
1,5
s
Fig. 7.12. Computed transients of DC motor with currentjspeed cascade control iIITorqet ü)o
\
úJRef \ /
o
úJ
~
I'
"' 1,.,; . 7 . i I . 't ll l., ' ,01 d" " ilv' ',," l l.n
Computed transients of a DC drive with currentjspeed cascade control are seen in Fig. 7.12. The parameters assumed correspond to those of a thyristor fed drive with a six-pulse-converter, Chapo 8. The controllers have been de signed as outlined in the preceding paragraph; if a rotating motor-generator were used instead of the power electronic converter, the response would nat urally lw llllldl slower, particular1y in the current loop. AI. Jirsl., a sLarl,illg (,ransient is displayed, caused by a step of the speed ref ("ICll(',(~ 1'J'll/1I '/,!'l'n Lo 0 .:\ Wo. Tltc p{r( ~d of CIl1T<'lIt, limit dllring the acccleration ph;I::" ;,: : \\".Ii /Iii /.I", dyll:l.Illic ("olll.!'()l erro!' are c1(~ar-ly :;('(~II. '1'11 i:; I.rall::icIlL
!l()
7. Control of a SeparateIy Excited DC Machine
7.3 Cascade Control of DC Motor in the FieId-weakening Region
corresponds roughly to the transition of the operating point in Fig. 7.10 from () lo 3 via 1, 2. At time t = 1 s the drive, having attained the no-load point (:1) , is loaded with nominal torque. This causes a transient speed drop but the drive remains in the linear control regime, eventually reaching the operating poillt 4. Prom Fig. 7.12 it is seen that the source voltage e a rises with the induced vo ltage, which is proportional to speed; in addition there is the resistive and luductive voltage drop in the armature circuito The principie of cascade control has proven its usefulness over the years III ill11umerable applications; as mentioned before, the two leveis of control III ay be extended in both directions, for example by adding acceleration or position controi. Since cascade control is by no means restricted to DC drives, W( ~ will come across more examples where this exceedingly fl.exible method of mlltrol can be used to great advantage, see also Chapo 12.
7.~
Cascade Control of DC Motor iII the Field-weakening Region III S(~ct.. 5.3 it was shown that by reducing the main field, a DC motor can II(" ()p(~rated above base speed. For good utilisation it is important that the
111"tor should be supplied either with maximum field- and reduced armature voll,ag<' or with maximum armature voltage and reduced field voltage; as 111('lll.iOIl(~d uefore there can be some overlap for convenience of controi. Si11('(' Lhe motor may alternately operate in either one of these regions it is :"l'proJlrial.c to choose a control strategy which fulfills both requirements and I "'1"111 iI.s a continuous and automatic transition from one operating regime to
with a constant reference, e.g. lema", I ~ 0.9 Uao; the e-controller then attempts to remove the difference by adjusting the field voltage. Since, however, the reference voltage is out of reach as long as the motor operates below base speed, the controller containing an integral term will be saturated and the field power supply produces maximum field voltage which is the aim in the armature control range. When the speed is rising, the induced voltage lei approaches the reference value lema",l; after a balance has been reached, the e-controller leaves saturation and begins to lower the field voltage to limit the induced voltage, thus automatically weakening the field as desired. Of course, the initiative for a change in field voltage must always come from the armature because this is the direct channel of the speed controller, working through the armature current loop. The arrangement drawn in Fig. 7.13 represents a complex and highly nonlinear control system, confounding all general design methods; however, considerable simplification results if one succeeds in speeding up the e-controI loop, active in the field weakening range only, so that e may be approximated in that operating region by a constant impressed voltage. Such an approxima tion is not entirely unrealistic if the field power is supplied from a converter having a sufficiently large ceiling voltage (only one polarity of the field cur rent is required). If the stator and the field poles are laminated, the "control plant" may be represented by a simple lag element, having a speed-dependent gain, and the fast dynamics of the actuator; no serious stability problems are mL
mo
( '
11. "
H.. i"
di"
(7.1 I)
" " -/ ll.. ;
III .'I.II II,I"!',II' · (.,,[('1 III I,/ II/" "i l'('flit, 1101. fl hoWII III FI'" ./ 1:1, 1I 111.'y 1)(' 1'(11 1'"rJ" . " I ' l d tl fl ll'. tHd "II I,'.'VI·I'/dll/,. dl l VtlI'I l ill. l 'I 11111, 111111 ,(,1,011
11I1I'!!
foI' 1.1 IiI-<
Iii l 'IIIII JllIl','d
Fi/ter
ú)
%
1.11<' "l.IwJ'.
/\ (', nll.rol scheme that has proved useful in practice, is shown in Fig. 7.13 III ,: illlplificd form [G20, V9]. It contains the currentjspeed cascade control 11<1. i I1/'; through the armature voltage, as discussed before; in addition, there III 11,11 ;l.Il xiliary controlloop with the field voltage as actuating variable, which Ims I.IIC ~ purpose of limiting the induced armature voltage as the speed rises I)(,y olld base speed. The adjustable voltage source supplying the field winding, iei IIS II;dly realísed by a thyristor converter, which is represented in Fig. 7.13 I'y a la!,!; cl(~lnent. The nonlinear model of the magnctic circuit corresponds !.o Io'i/(. ;).4. /\ r(~cdklclc signal for the induced volt age CilU II(' n !c()l1structed from the 111( ';I.SlIl'cd annature voltage and current accoJ'( lill t.( 1,<) 1,11<: vo lta,gc equa tion
91
Ú)Re'
Wõ
Limiting
1el UaO
I emr1. 1--o-..t
ullo
- control/er.
1"IK. 7 . 1:1. ( :,,111,,'," "i' I )(: II "'Lo I' iII (.lU ! IIfIlLl.LlIre a.lld fidd wP',d{(:ning
regiolls
7.4 Supplying a DC Motor from a Rotating Generator
7. Control of a Separately Excited DC Machine
~) '2
.!!!...j mM
be expected even at high controUer gain. Neither is the nonlinearity of I.he magnetic circuit of great concern since the e-controUer operates predomi lI alltly at reduced flux where saturation is not pronounced. For these reasons a PI-controUer is quite adequatej its lead time may be tuned to the field I.i Ille constant of the motor while the residual delays are taken care of by the r<'llIaining I-term of the controUer. Assuming fast e-control in the field weakening range, the simplified equiv al( 'llt block diagram in Fig.7.14 results. There the effect of the e-controlloop (:; :l)lproximated by a constant volt age le max I acting like an external distur h;l.Il ce and the flux approximately becomes a nonlinear algebraic function of 1.0
~W 1,010,~ ~ Wõ
_W
i'';;';/J
II
I l
1
mo
WRef
L/Wa __
_____ >
~--
ia -ç
lPe ~eo
1,0 0,5
-I
Wo
'[>eO
r-
I'
lPe lP ea ~
mo
úJo
: :p('('d,
1> e
for
1:01> 1,
allccLing the mechanical time constant. ln principie, this undesirable change
"r pa rameter could be compensated by dividing the speed error by Pe/PeO I.he input of the speed controUer, as is indicated in Fig. 7.14j however, in III ()S(. cases this is omitted and practical results prove that the correction is II s llaIJy not necessary. Some consideration must of course be given that the :, pe('c! COlltroUer yields acceptable results in the armature as weU as the field rl/lIl.rol regime. As an example, Fig. 7.15 shows a simulated no-Ioad reversing I.r; LIIs i(~J\t between ±1.5 Wo that was computed on the basis of the system :: I! " WII in Fig. 7.13. The effect of the e-control loop is clearly recognisable 1', ()1I1 III<' ">,.- and the e-traces. 1I. h ; I,,~ IJC(m mentioned that the use of a fast e-control loop with high r"r" il lj ': voll.age across the field winding calls for a laminated motor frame. W 11.1, b"I':' ~ Illotors it may be necessary to employ a more accurate model of LI ... li,'ld ,'jrclli t. for the design ofthe e-controlloop, taking eddy currents in the II "" iII!." :ICCOllut [G20] . AIso the high voltages induced by transformer action III LlI<' ;1.1"I1I:t!.nre winding affecting the commutator must be considered. To
:,!I- --
:d.
u. M()
~ iao
.!!§...
{1 1, r,.,
93
!!!.b.. mo
- - -ç ia' - -
'r
~Reversal of
ea
e
:
speed reference
uao' UaO
1,0
--,
~
~:-"'~uao , :-...
, :-...
;::... ~
uao
;::...
;::...
;::...
:-...
Fig. 7.15. Computed reversing transient of speed control system in Fig. 7.13
eliminate the effects of winding temperature and supply voltage fluctuations on the main flux in the base speed region, an inner field current controlloop is often included which maintains a prescribed maximum field current in the armature control range independent of a changing resistance of the field winding and helps to speed up flux contro!.
~,:\~}
(110
,.
«/'"
<1111/ 1
( 1./111
I 't'l ll 1( '11<. 1/'
emexl . (ü) l-uao 'slgn "'o 1\ / ÍlllilO!
7.4 Supplying a De Motor
from a Rotating Generator
,
((uel a ui II/ui cI
cOl/lrol
III /lul< I WII. 1/" " li" li I "U;OIl
I" IH . 7 . 1'1. Sl lIq dW"d " "1>1 " ' /(' 11 1.' 1.1.;1111 "l' nl', \(,,1 t'" " (.If .I I, ,, ,1' ill l i"loI WI\lI l1l' " I'''I ', "/l1l11"
lu HlI~ prpc(~c(lilll~ sections was mentioned that any adjustable voltage source (".ol1ld :,\('\'V( ' ; \:; p(Jw( ~r :->npply for the armatllre of the De motor. The classical ::('11('1111' (Ir I )( ~ III ' d.or ('o\l!.rol (kvdopcd hy W:I,rd T.(~ollard [L21, 1,22, Y 4, Y51 ,
H,I
7.4 Supplying a DC Motor from a Rotating Generator
7. Control of a Separately Excited DC Machine
1'111 pi oys for this purpose a rotating D C generator, the armature volt age of wllidl is contro11ed through the field. Combined with up-to-date control, d ri VI~S of this type are still in use today. The basic circuit is seen in Fig. '/ . 1(;; Lbe separately excited DC motor (2) driving the load, for instance a rullillg mill or an elevator, is fed from a separately excited DC generator, J"lllIlIillg at approximately constant speed. While a rolling mill or a gear-Iess "I"vaLor drive would norma11y operate at very low speed, for example between I 100 .L/min, resulting in high torque and a relatively large frame size, the 1'.'·llI'raLor being of similar power rating as the motor can run at any conve Uí' ·lll.ly higher speed Wl, with the aim ofreducing its size and cost. lnstead of :1. I l(; gCllerator, an AC generator with rectifier could be used for supplying Lili' I)C-link; of course, this would rule out regeneration. 'I 'lte prime mover of the rotating converter is most frequently an AC induc Lil III I.llotor (1) with cage or wound rotor, Chapo 10, requiring no continuous C"IlIII.rol. The power fiow of the drive is then reversible with the contro11ed Illut,or being able to operate in a11 quadrants of the torque/speed plane. ln ("as. ~ of a ship- or vehicle-propulsion scheme a turbine or a Diesel engine could bc sllhstituted as prime mover (1) (Turbo- or Diesel-electric drive), while the drive lllotor (2) would most likely be a series type motor. Resistive braking cUllld he employed instead of regeneration which is of course not possible wil.lt a llllidirectional prime mover. W iLll largc rolling mill- or mine hoist-drives the total drive is often split IIII iIILI I several machines; for instance, on a large reversing mill, two twin 111()l.ur~ Illay drive the upper and lower ro11s without direct mechanical con 11 ...·l.iUII :l.p
nrLiC1llar interest with weak supply systems but its importance re cl'dceI ;,IS LlII~ slIpplips !lave become stronger anel more closely interconnected. (:ollvI ·rsl'i.v, ;1, J'()l.nl.illl~ COllvcrtc~r can servI: for protl~(".t.ion of sensitivc loads, i i i Ir" ;1:: [l il.pl'l' IllilLdriVI':i, agaillSI. Short-l.IIIlC lillO di sturhallccs. Silllilar argll 1111 \111..'1 Ilrn vl tI,,1 wlll'll 1I111111.1'ITllpl.ildl· [lOWI'J' rol' hllplirl.alll loat!s is 1'(~'111i['(.'d,
95
for example computers, elevators or operating rooms in hospitals; with an intermediate motor-generator-set supplying these loads, there is more time for switching to an alternative supply in case of a line disturbance. The main drawback of the Ward-Leonard-drive is of course that, besides the adjustable speed drive, two additional machines of about the sarne power rating are needed, Even though they may be smaller in size because of higher speed, large machines will still require heavy foundations and continuous service of bearings and commutators. It is for these reasons that static power converters, particularly of the modern semiconductor type have supplanted rotating motor-generator-sets in new insta11ations. If the field of a rotating generator is fed from a rotating exciter, both responding relatively slowly, it is with some care quite possible to control the drive motor manually and without any automatic equipment. However, the machine cannot be fu11y utilised because of the danger of overcurrent; feed back control with its much faster response then offers considerable advantages making it possible to take various side conditions into account so that the AG supp/y
Rotating converter set
•
DG- generator Induction I Motor 1
DG-motor
Load
III I!
'
Starting resistor
~~
-
.o~.~~,--------------
---t
Fi,l. 7.111. (~I)III.I' 1)1
e max
lil'ill'llIl: ()[
WMel L(~OIlIl.rd · drivc:
!II'
7. ContraI of a SeparateIy Excited DC Machine
drive assumes well defined operating characteristics and may be inserted into n,lIl.olllatic production processes. By exciting the generator and the drive mo 1,,1 I.ltrough static converters with high ceiling volt age (2 to 3 times steady ::Iill.(' voltage) the speed of response of the drive may be sufficiently improved 1m Illeding most dynamic specifications. 'l'lw control structure shown in Fig. 7.16 is essentially in line with that in I"il': · 7.13. The only difference is the possible introduction of an inner control loop ror the armature voltage U a of the generator [876]. 8ince the pertinent 1'()Jlioll of the control plant comprises only the field time constant of the 1',('ll<'ral,or and the residual delay of the electronic field power supply, the re ::pOIISC of this control loop can be made quite fasto Its advantages are the lill('ari,sation of the inner part of the plant, the reduction of the effective ar 111:\llIr(' impedance of the generator and the possibility of precisely limiting 1.11<' ;tIInature voltage. This is important with large machines where the per III issi ble limits of commutation must be fully used without running the risk 01" cx('cssively sparking brushes. For the superimposed current controller the c!os('d voltage controlloop looks like a linear actuator with constant gain and (·oll.si(lcrably reduced lago
8. Static Converter as a Power Actuator for DC Drives
8.1 Electronic Switching Devices A characteristic feature of the rotating converter discussed in 8ect. 7.4 is the consecutive power conversion from electrical (constant AC line volt age) to mechanical (speed of motor-generator) and back to electrical form (variable direct voltage) from which it is eventually transformed by the drive motor into controlled mechanical power. These conversions constitute the advantages of the control scheme (separate electrical circuits, decoupling by rotating masses) as well as its draw backs (cost of machines, foundations, power losses, servicing, limited speed of response). This naturally gave rise to early ideas for eliminating the intermediate mechanical stage by supplying the DC drive motor directly from the AC grid through static electronic devices. That such possibilities exist in the form of static switching converters has been known for a long time [A15,W29,46]. ln fact, switching devices are the only available means for high power converters because any other circuit components would result in prohibitive power losses. The basic element of a power converter is an electronic switch having the properties of a controllable rectifierj the idealised steady state characteristics of different switches are seen in Fig. 8.1. An uncontrolled diode rectifier, Fig. 8.1 a, has two distinct states: With reverse volt age (u < O), there is only negligible leakage current, whereas in forward direction (i > O) a low forward resistance combined with a threshold voltage exists. Controlled electronic switches have at least three different states: A thyristor, Fig. 8.1 b, is blocking reverse current like a diode, but in forward direction (u, i > O) two possible states exist: the device may either be "conducting" 01' "blocking". The first state is again associated with a low conducting resistance, typically 10- 3 [l, while in blocking condition the device exhibits a very high blocking resistance, for example 10 5 [l, similar to that in the reverse direction. Hence the thyristor valve may be approximately described by a mechanical switch in series with an uncontrolled rectifier (diode). The main difference between a thyristor and a diodc is tlw existence of the forward blocking state. 'flw I.rall:iitioll frorn "blocking" to "conducting" may be initiated by a slIlall 1'::\,1<' ('111'1'<:111. plllse -ii" while the inverse transition fl"om "conducing" lo "lllo('liilll'," (' il,II:.i for a :/'('1"0 inl.c:rscci.ioll of UI(' 1I1aill cnl"n:llt, "Tnl"ll-0fF aI. ('111'1', '111 '/,(\("" " ', I. hyr l::I,llr:, a r, ' wid('l y 11:; ('<1 ror ('olil,rollillJ': U (:
'!U
as a Power Actuator for DC Dríves
H. ,':iUtLic Converter
8.1 Electronic Switching Devices
n~cent
development are Gate-Turn-Off thyristors (GTO), where electronic blocking is also possible, Fig.8.1 Cj they are mainly Il ro , -c! III rorcc-commutated converter circuits with a DC link, to be discussed III ( :lld,P: ;. ~).2 and 11. Other types of electronic switches which are of interest ',,1 ::,,,,ci;d applications are reverse conducting thyristors or voltage controlled 1.1 1I. 1I :IIKl.ors with internal diodes, where the two forward states are combined IV iI,11 I('V" ['sc conductionj an example is the "lnsulated gate bipolar transis 1., 'I" (I ( : I ~T), seen in Fig. 8.1 dj there are also applications for symmetrical 11 11' 11."1 il' S, l"ig. 8.1 e, as wiU be shown in Chapo 11.4. fi
11101('
1',:11,,' '''Idrolled
forward conducting
conducting
,
u
IUverse Iilucking ,I)
conducting
!./ ",electranic rückwarts sperrend
I
\ control
forward u blocking
b)
",
/ ' " electronic \ control
!ig
u
o
Oiode
! f>r
blocking
reverse blocking c)
. I
o
o •
-u
~
!ig
f>(
o
-u
Control/ed rectifier Thyristor
Gate tum·of( Thyristor (GTO)
conducting
/ ~. JI , VO / !Jt l
" " '/Iludiu!}
"J
, conducting
~"
electronic \ control
forWârdU blocking
u
u
Insulaled gale bipolar Transistor (IGBT)
... l,~ .
H.
\
e)
electronic contraI
:. . ... .'. . .. .
blocking u
electronic \ control"-.., conducting
-1 \f.
~
,
blocking
1
i
LfJt
Ug2
--.J \.
/
L
~ u Symmetrical switch (IGBT)
I. I)II COII trolleel anel controlJed electroni c switdlC::;
III I.he pax(. Lhe fllllct.ion of controlled d(~cl.J'()lIi (' x w;Ldws lias been realised with vm;"lIs dl'VÚ'('S Ill,tkiug \l*~ o[ diJkre eurly Ilrtllll,ioll: ; nll' 1I1l/l11kf,(' (.odny. Siul't: l'I(\('.(.I'OIl Llrl1oc 'lI wít.1t Itl.'i t!.l:d· cal.hodl~s cOllld 1I"t. IllIpply "II"II I \ IJ l ' IIIT"IIt. 1,0 drivo IIIOt.OHI II JltI \\11 ' 1< ' pl nl~ll<'d wil,lr lri((1r voll, 'I '," cll"I" I " 111 ri I,! I,!', III 1""11 d ll d l'll" Y, 1.111',1' W "I CI! , III""IIIIt'C("d I,,Y J'~ 1U 1 lilll ..,c)I.II I'('I!
99
with heated cathode, called thyratronsj however, power rating, life time and reliability were also rather limited so that they were only useful for small drives. High power in the MW-range, such as needed for rolling mills or mine hoists, could only be supplied by mercury-arc valves having a liquid mercury cathode. These electronic valves were developed to high perfection by the 1950's but the liquid Hg-cathode limited their use in vehicles and remained a disadvantage, as well as the large size and the relatively high voltage drop across the electric arc connecting cathode and anode (20-25 V). Outside a narrow operating temperature range, controlled mercury-arc valves showed a tendency to occasional spontaneous arc-backs, i.e. conduction in reverse di rection resulting in short circuits and high over- currents. This required close temperature control by heating or cooling the steel vessels. AIso, mercury arc converters needed relatively long intervaIs for deionisation which made them unsuitable for operation at higher frequency. After a brief interlude with magnetic amplifiers where magnetic materiaIs with an approximately "square" hysteresis loop were used as "switches", alI these difficulties were completely overcome with semiconductor switching devi ces (Transistor 1948, Thyristor 1957) which have become available at steadily increasing power and with improved performance since about 1960. They have replaced mercury arc valves even at the highest power ratings in high volt age DC transmissions (HVDC). The physics of a thyristor are extensively dealt with in the literature, e.g. [17,24], a simple description may suffice here. ln Fig. 8.2 a a small slice of highly purified monocrystalline silicon having four electrons in the valence band is shown, which is doped with materiaIs of valence three or five, for instance Boron or Phosphorus, so that three P - n junctions are created, separating the layers of P - and n - conducting material. The inner layers are relatively weakly doped while the outer ones, forming the anode and cathode terminaIs, contain more doping materialj the inner P2 layer is also connected to a gating electrode supplied by a low impedance control circuito The following steady state operating conditions exist: • With negative voltage between anode and cathode, u < O, the diode junc tion PI - nl is reversely biased; this corresponds to the reverse blocking state seen in Fig. 8.1 b. • If positive voltage is applied (u > O) but without gate current (u g < O), the diode nl - P2 has reverse bias which constitutes the state of forward blocking. Again, only negligible current flows in the load circuito • With positive voltage, u > O, and with a sufficiently large gate current, ig > O, the junction nl - P2 becomes conducting, with the load current i satnrating the region nl from both sides with charge carriers. Thus the lhyri::; tor assumes very low resistance in forward direction. When i has l')( ('e(~(kd ;t cl'/'I.ailllatching threshold iL, the conducting state persists even afl.e/' Lh e' (',;tl,( ' (', lIl'rmlt: is removed .
I1I1I
H,
8.1 Electronic Switching Devices
Sl.atic Converter as a Power Actuator for DC Drives
(hll .\' Ily )'('dllcing the load current i below a lower holding threshold iH can 1.11<' I.hyrist.or again revert to blocking statej hence it represents a switch that ,'11. 11 1>(' ('Ios(~d by a low power control signal, whereas opening it requires the 111 11 <1 ('lllr(~l\t to become very small or zero.
u 1\
A77
200A ,1,5VI conducting
ig
u
K
"
101
i
b
.
/u9 K
iH
-1kV
= 100mA
Ig
forward blocking
reverse blocking
G
1kV U
Fil,(. H.2. Thyristor model and symbol
Fig, 8.3. Static and dynamic characteristics of thyristor
:-;illlplified characteristics of a typical power thyristor are shown in Fig. ,"{ .:\; Lhe scales of the curves are different to emphasise the details. While the kak iW;I' cnrrent in both directions could be below 100 mA, the load current in l'lllldllcting state may be several thousand A. The voltage drop of the (,()lIdlld.illg thyl'Ístor is about 1.5 V and the maximum volt age in blocking ::I.:Ii.I 'S call Iw lllore than 8 kV. W!II'II tllC lll O) the effects are 111111" 1'lllIljllt-X, Whcn the voltage reaches a breakover levei, which depends 1111 1.11<' I',ii 1.1' (,\Irr(~lll., the thyristor switches spontaneously to the conducting : ,1.:1i." 11,,\V('vI')' , 1.11I'l1-OIl by increased forward blocking volt age is not a suitable 1111"1,, "I' (' IIliI.)'o! for a power thyristor because the "firing instant" (a relict I" 'III pla::lll i(, discharge valves) is not sufficiently well defined and there may I", 111(':11 Ilvel'iwal.iug of the semiconductor elemento lnstead, the switching L" LI", ()rI " sl.ill.<: should always be effected by a gate current pulse having a ::1" ,rI. 11:;1' I.ílll(' which has the additional advantage that the thyristor fires at a 1"1 '(' itl,,1y ddiued instant, independent of the individual thyristor parameters. 11i)(' 1.0 lhe small dimensions of the active semiconductor wafer and its III \V 1.I1I~l'Il1a! capacity, the safe use of thyristors calls for observing some re ::l.lldiIlIlS; apart, from limits on temperature, volt age alld current there are IlIil,Xi11IIIlll values for the rate of the current rise (di/ dt) while switching On, as \1'1 ,II i l.~; UIC voltage rate (du/ dt) during and aft( ~r tu rll· OIf. This requires the ill('III: :il)ll 01' prol.cctivl; circuitry such as a seri.::; illdll('I.i\.IlC( ~ iUld a parallel ca JI ,)('iL( II' (.~llIlbl)(~r), COllsi(kring al:-;o tlte swil.chil'i ~ Hllq';<':, t.llat. i\.l'e llu
recommendations by the supplier are observed, the thyristor has proved to be a reliable and very robust switching element that can stand high short time over-Ioad current, such as a 10 fold surge for one cyde. Persistent overload must be prevented by a fast acting closed loop control and in the limit by fuselinks. An important design parameter of thyristors, particularly when applied in force commutated converters, is the recovery time tr· It indicates the time that must elaps e from the instant when the load current has reached zero, until the thyristor has attained full blocking capability in forward direction (u > O). The purpose of this interval is to gi ve the residual charge carriers accumulated in region nl during the conduction period time to decay through diffusion and recombinationj depending on the type of thyristor, reverse volt age and junction temperature, the recovery time ranges from a few Jls for a high frequency thyristor to several hundred Jls for a high power thyristor designed for low voltage drop. t> tr / /
sfored charge
I, .... _ /
I I
I
r::.:J u
/u(t)
o
i
\.
a Fig. liA. '1'lIrlld lfr I.mllHil~nt of thyristor (n) (:1111"111. i'.lId v"It.II.I~" (h) I'roLeclive circllil
I-
b
...I'~_I .. I
o
I I ): ~
K.
8.2 Line-commutated Converter in Single-phase Bridge Connection
SI.;tLic Converter as a Power Actuator for DC Drives
fi Lypi cal transient of the volt age u(t) and the current i(t) of a thyristor dlllilll ': Lllrn-Off is depicted in Fig. 8.4 a. As the load current i reaches zero, 1,11\" I.hyri stor remains conducting for a short period with reverse current, thus (' 1 ("; Llill ~ the charge-carriers from the junction. After this charge has disap P( '; II ('
Line-commutated Converter iII Single-phase Bridge Connection
H.2
'I'ilyrisl.ors, which are available in a variety of voltage and current ratings, ;1./ (' e'Jlllhillecl to converter circuits to serve as highly efficient electronic power :: lIl'l'li('s for a wide power range. Of particular importance for the supply of 1)( : IIIOl.orS are the so called line-commutated converter circuits fed from a ::iUI',11' o)' Lhree-phase AC supply of constant volt age and frequency. They are '1I Iil.:d,I" ror sllpplying adjustable direct voltage and current to the armature 1.1111 li ('ld willcliIlg of the motor, either directly or through a transformer.
LA
a
.--o
2'
I
io
Uo l' fi
iI
F l,( . tl .li .
(I\')
d ' ~ ' III I,
:-; ilfl \~ "
!o
iA
UAi
I
~ tJ
111;1,1,,:\' l, i. l w" "" II I1H!lt,at. ~· d ("O IlV I' rf,I11' (I.) l'II I,,'I:IIQ01 1',, 1 I''I" 'I, v,.I,' 1I 1.
io ."
i Uo
103
The single-phase circuits discussed first are of interest, apart from railway traction drives, for low power applications up to a few kW. The circuit shown in Fig. 8.5 a contains four thyristors, which are made conducting or "fired" in diagonal pairs during every alternate half cyele of the line voltage UA· Provided there is a continuous direct current iD, the function of the circuit may be described by a mechanical commutating switch, connecting the DC circuit with alternate polarity to the line voltage
UA = UA sinT,
wt =
(8.1)
T.
When assuming at first that the direct current is smooth, iD = ID, being maintained by a fictitious current source, and that the impedance of the AC circuit is negligible, with UA coming from an ideal volt age source, the volt age UD(T) between the DC terminals consists of periodic sections of the AC voltage or its inverse; this is seen in Fig. 8.6 for different firing angles a, which determine the instant at which the other diagonal pair of thyristors is made conducting. The firing angle represents the switching delay with respect to a "natural" firing instant that would be valid for a circuit consisting of diodes; for the thyristors 1, l' the "natural" firing instants are T = O, 27r, ... when the volt age UA (T) and - with thyristors 2, 2' conducting - the voltages UTl (T) , UTI' (T) become positive. The maximum firing delay is ama", = 7r; for a > 7r, the thyristors 1, l' cannot be made conducting since they are reversely biased. This is also seen from the curve of the volt age across the thyristor, UTI (T), drawn in Fig. 8.6 c for different firing angles a. Later the range of the control angle will have to be further narrowed. The trace of the thyristor current iTl (T), also shown in Fig. 8.6 c, elearly indicates the switching action of the thyristors; at any time at least one of the quantities UTl (T), iTl (T) is zero. ln steady state condition with a = consto the volt ages and currents in thyristors 2, 2' pertaining to the opposite diagonal pair are identical, being shifted by a half period. Finally, in Fig. 8.6 d the input current iA(T) for the three different values of the firing angle a is drawn. As a consequence of the periodic commuta tion between the two pairs of switches the impressed continuous current ID appears at the line side of the converter as a square wave alternating current iA(T), whose phase lag ep with respect to the line volt age UA(T) is prescribed by the firing delay, ep = a. The active power (mean of the instantaneous power) at the line side is - assuming sinusoidalline volt age determined by the fundamental component iAl (T) of the line current
PAC
= UA IAI
cosep
2V2 =- UA ID 7r
cosa,
(8.2)
whe['(~ (J II, I II' are H.MS-values of the line voltage and the fundamental cur n:lll. ('.OIIlj>OIlC·I\I ,. For LI < a < ~ , the power flows from the AC - to the DC :;id", wilil., ("r :: . 11 ' n""",,, (./1I~ (lO WN f10w is ITv( ~ rs( : d.
104
8. Static Converter as a Power Actuator for DC Drives
8.2 Line-commutated Converter in Single-phase Bridge Connection
105
rot
a
Inverter I Rectifier
IX
1
=
lX.2= 90°
30°
a=7C
(X3=150°
2 Fig. 8.7. Phasor diagram of a single phase converter
rot Uo
blocking voltage V,T > O is created for a < O, during which the thyristor can be fired. Forced commutation calls for additional components and thyristors with short recovery time which involves increased cost and complexity; this wiU be discussed later.
2.
rot
Üo
-u 11: Reclifier
c lX.2
«( I
'II I
I-----
.
~ ~ /II
_'AI \
/
I
- , 1\ Ir
,
I
I
rot
I
~ ~
I
7C
I
\ I
\ ~
rol
~Tr
I
'_I
,. ~Al . ----T-J IA
I
I
\
I
I
iA \
I
I
a
a3
,~
Inverter
\
t7' \
/
d 1"'1(' H.G. V()ltages and currents of an ideal single phase converter
'I'II(~
first mode corresponds to controlled rectifier-, the other to inverter
"llCral.iof! . 111 Fig. 8.7 a phasor diagram is showing the phasors of the line volt !l~" I J A
aud of the fundamental component of the line current,
I Al .
Clearly,
a cllul.roUcd converter appears as an inductive load at the line side, drawing Ja.j':I':ill p; J'( ~ a("t.iv(! cunent; this is due to the control principIe allowing only ,kla.y ( ~ d lirill,~.
Fig. 8.8. Control curve of phase controlled converter
The possibility of reversing the power flow by electronic control is of importance if 'the converter is to suppIy a DC motor since it permits regen eration. It is pointed out, however, that firing controI affects not onIy the active but also the reactive power flow at the line side. At a = ~ the reactive power assumes a maximum value (for I D = const.) while the active power is zero; this is the case when the motor produces torque at standstill. Identical results are obtained when examining the DC side of the con verter, where the mean voltage is
Firiup; adv;mce is uot possible with line-commutatcd convcrt
1H".ctUl s(' Lhe jiriug plllses would bc appli(~d (.0 tltyristors having rcver s (~ Il ta!-l volL i'lt'í< ' whirll i'f('v('uLH tlH:lll [('IIJH (·OJldlld.illg; advaúC(~ firillg i R ollly !,ow d ld(' wil.h· r"I" '" ,·o)[)l ll u(.n.I.('d '~OIlV .. [ I.'· lii , wI"," : illI iul.<'rval wil,h rorw
!'I ii
........
\
!
o+rr
tt u
1
71'
'It A
sin T
2
dT
= -7r í/, A
cos a
2)2
= 7r-
UA cos a
= UDO
COS
a
UU)
fi I U I I
I
,~
C f\
8. Static Converter as a Power Actuator for DC Drives
1.06
8.2 Line-commutated Converter in Single-phase Bridge Connection
This is described by the control characteristic UD(Ú) seen in Fig. 8.8; for reasons discussed later the curve is only valid in the interval O < Ú < ú max , with Ú max ~ 150°. Because of i D = I D = const., the mean of the output jJower is F DC
2..)2
= UD I D = -7r- UA I D
cos Ú
(8.4)
,
which is in agreement with Eq. (8.2). This is so because the idealised switch iii! ; converter consumes no power; it functions by shifting the 180 0 -blocks
()r Lh e direct
current with respect to the line voltage. The transferred mean changes sign in accordance with the direct voltage as the firing angle II" exceeds i. Since the current cannot change sign because of the unidirec Liolla[ thyristors, the line-commutated converter circuit in Fig. 8.5 represents ;t cO l1 trolled power supply operable in the two quadrants of the ID, uD-plane, ; lS seen in Fig. 8.9. The control is performed by electronically delaying the li rill f.!; pulses to the thyristors by the firing angle ú. p.. wcr
'.
107
Controlling a converter by electronic phase shift of firing pulses requires little control power and involves only very short delay. This is seen in Fig. 8.10 showing the output voltage uD(r) with the firing angle being switched between Ú = 30° and Ú = 150°. Since only one diagonal pair of thyristors can be fired in each half period, there is a maximum delay of 10 ms with a 50 Hz supply until the converter output responds to a command calling for a chang e in firing angle. Hence a static converter represents an actuator of high dynamic performance combining - high power gain - large output power and low los ses - very fast response . These features may yet be enhanced by employing multi-phase circuits, as will be shown in the next section.
(Ut = r
Uo
"
'/1' (//1
a=O
Fig. 8.10. Output volt age of single phase converter with changing firing angle
Rectifier
lO Inverter
a= amax /I
1//1
I"il~ · 1'1.0. ' I'wo
quadrant operation of line-commutated converter
(;Icarly, a switched converter causes interactions with the line, involving reactive power as well as current harmonics. Because of the supply / 1,11(1 1.11(' Lrausformer leakage impedances this causes voltage fluctuations and hlu'IJlollic ([i stortion which must be taken into consideration as the rating of 1,1", ('()Ilv(~rt(']" inereases. While the harmonics problem may be relieved by I',(JIII, ~ t,o Illlllt,i-phasp. circuit,s, the reactive power is a characteristic feature of alll ill( '-cOllllllllt,al( :d CO llverter::; with ddayccl firill g control; it is independent of LI", 11111111)('1' 01' pha~.,\,s alld call ollly 1)(' (:(lllllLt'racl,cd hy reac tive compou:mtioJl, AI. ili!'.h pmv"1 II()i.h ,,11" '1'1.:1 IIln.v J'( ~qll, ij'(' C"l'I'l'djoll I,y hi\.J'IIl()I;icJill,, · r~i havillJ>; II, I"'" dill!,, IIII'IIL ,""V"I' 1'111'1 ,01' u.t. lill( ' 1''' ''111 1\11' '.1' I;I./',I ~ illg
One drawback, though of little practical importance, is the fact that the steady-state and the dynamic behaviour of the converter is highly nonlinear, corresponding to a nonlinear sampled data system with one firing transient in each half period, From Fig. 8.10 it is apparent that, following a changing firing command, the result may not become effective immediately because there are variable waiting intervals. This is particularly pronounced when the firing angle is increased for inverter operation because it is necessary to allow for the relatively "slow" reversal of the line voltage. When going to rectifier operation, i.e. reducing the firing angle, the converter may re spond instantaneously. An accurate analysis of dynamic converter operation is quite complex; for small perturbations linearised difference equations can be derived but large signal dynamics are governed by non-linear difference equations for which there is no analytical solution [F12,L34,P46]. Fortunately the converter is usually followed by an inductive load having low-pass transfer characteristics, such as the armature circuit of a DC machine which smoothes the current; therefore the converter dynamics may be approximated by very simple models when designing the inner current controlloop. Anot,lwr Jloint where high power converters may occasionally cause prob kws is a djj'(~ct. collscquence of its advantages mentioned before; the power HIIJ)pli, :cI Lo LIli' lontl IIlw;t, iustallt arwollsly COIIIO frOITl the AC líne. There is II I) d' ·(O""plllll'. 1'1(" '( I. t,.y illt,I ' l'llaJl y sl,( IITt! ( ' IlC'1') ~'v hl·I ':\.II S(· ali id('al C() ll v(~ r" ,(~ r
lOS
("ontains only loss-free switches. As was mentioned before, the importance of Lhese effects has receded as the capacity of the interconnected supply systems Itas increased and is more closely meshed. The control of the converter is normally performed by the output signal I)r an electronic controller influencing the phase of the firing pulses through ii, firing control device. Its basic function is explained with the help of Fig. !:UI. The output signal Yl (t) of the innermost controller, usually the current coutroller, is compared with a saw-tooth-shaped periodic ramp function YT which is synchronised by the zero intersections of the line voltage UA. When eVI'r the difference becomes positive for the first time in each half period, lhe diagonal pair of thyristors, Fig. 8.5, having forward bias voltage is fired. '!'/te selection is done by alternately activating one of two monostable trigger circllits MT 1 , MT 2 . One ofthe trigger circuits produces a short pulse which, ;IJt,I~r amplification, is passed through a small isolating transformer to the 1~"Lcs of the thyristors to be fired (only one of the output windings is drawn). ILL high voltage applications, such as high voltage DC-transmission or static n:- Now a I.ltyriHL()r pai r 1::1,11
«
S.2 Line-commutated Converter in Single-phase Bridge Connection
S. Static Converter as a Power Actuator for DC Drives
(rom controller
2
109
.
Y1 =]iarc sm Y
rl~LL~~
~I~I: Y I;====IJ
T !.
Il0ug,
__ .
~IIOug2
a 0:1
0: 2
0:1
0:2
~
-Y1 (7:) wt=7:
b Fig. 8.11. Principie of an electronic firing control circuito
(a) Block diagram; (b) Mode of operation
UD~
Uo =
",. a(," \ /";{t !.. '
_ '.
I
'"
;'
l,
'. .'. ;.... . •
ROiO
I
(,o .
I \
I \
I \
I \
I
'.
~
/
I
I
1~ 1\ I~\ "
I
;"
I
F
\ \
I
_"">L-_. .
\
rot = r
\ \ \ \
,/
Fig. 8.12. Single phase converter with l'esistive load
only begin to conduct if the line volt age exceeds the back voltage, IUAI > e, because otherwise the thyristors would be reversely biased; the current ceases to flow as soon as this condition is violatedj Fig. 8.13 shows the effect. A furtlu:r cornplication arises, if the DC circuit contains an inductance J~ II in ,t () !lecami(: th(' arlllatlLre indllctance with its
UI)
8,2 Line-commutated Converter in Single-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives
110
8 ~
/
~
The solution, satisfying the initial condition iD(al) = O, and with the abbre viation LD/ R D = TD is
,
e
\
\ \ I
I I
iD(r)
\ \ \ \
I
,
I
rot = r
UA = --
1
\ VI + (wT D )2
RD
[sin(r - arctanwTD) -
I
e-(r-a,)!wTD
1"11 A 1
- .!!...- [1- e-(r-a')!'OTD] RD
energy can maintain the current iD during part of the interval where
< e. This is drawn in Fig. 8.14 for different values of the firing angle a
;tlld assuming e = E ~;I.ill neglected.
= const.. The internal impedance of the line voltage is
lO
/ I /1
J/
1
",
E
(8.6)
At f31 < aI + 7r the current iD becomes zero, iD(f31) = O, see Fig. 8.14, at which time the thyristors return to blocked condition; the current remains zero until r = aI + 7r, when the thyristors 2, 2' are fired and the sarne transient is repeated. Thus in steady-state the direct current is discontinuous, consisting of separate pulses. The mean of the current is a function of all pararneters, _ iD
'\.
sin(al - arctanwTD)]
/
1"ilJ:. 8.13. Single phase converter with back volt age and resistive load c;I.()II~d
111
1
= ;:
j{3'
iD(r) dr
= F[a,
UA, wTD , R D , E].
(8.7)
a,
I
~31a1
I
I ,I
III)
I' I
rl
I
I
I
/
--úA I' ...... __ ......
I
/
' I...1 ...... _
//
0.,+11"
A
UD =
iD(aI) = iD(al
:a1
2n rot = r
a2
,
rll
I /{ JI
,
1 /I
'
'U , \ HIII ,
/,/. 1
I'
~
/1 1
•
(I)};)
+ 7r)
= O;
(8.9)
thus it assumes the form UD = E
Discontinuous and continuous current
,b:"
(8.8)
This integral may be simplified with the help of Eq. (8.5) considering the periodic boundary condition
",..-'
115
J
uD(r) dr .
a,
I I
assume an initial zero current state, iD =diD/dt = O and conse = E , At time wt = r = aI, with the condition UA sin aI > E, 1,11<: I,hyrisl:ors 1, l' are fired so that the current iD begins to rise with finite :,Io(l(: and lIuickly exceeds the latching value iL. With the thyristors 1, l' ('olld udi II"; ;wd ;11, /I (1') 'Il A (r), the following diffcrential equation holds, {"
~ j
·1
(lll(~ntly '/Lo
r, I
When evaluating Eq. (8.5) the impedance (RD, LD) must be augmented by a smallline-side impedance as is shown lateI'. A similar expression holds for the mean voltage
-- U ......
I I
I
o
Lct
I
l I I I, I
I'
I '"
a
I"i!;. tl.14.
I
/
I
rot = r
~3+n
+ RDiD
(8.10)
where iD is given by Eq. (8.7). This load characteristic uD(iD) for a = const. and different values of the back volt age E is highly nonlinear as long as the current is discontinuous; this is due to the variable period of conduction, f31 - aI < 7r, as is seen in Fig. 8.14. When graduaJly reducing the firing angle, a < aI, with the same initial conditiolls as 1wfore, the amplitude and the duration of the current pulses illc\,(',as(: 1IIII.il :I, lilllitiug case occurs, a = a2, where one pair of thyristors in lir('d ;11, UI(' 111.111]('111. (,]1(' other pair ceases conducting, f32 = a2 + 7r; the ('1111"1'111. 1.11<'11 ollly t,i1lwlll':; 'I,('II~ OIlC(~ (:v('ry IIn,lr p(:riod.
8,2 Line-commut~ted Converter in Single-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives
112
If the firing angle is still further reduced, a < a2, the current will not reached zero by the time the other thyristors are fired. This then con :;LiLntes the region of continuous conduction, where the current is transferred iII a commutation transient directly from the outgoing thyristors 2, 2' to the illCOllling thyristors 1, l' and the current is no longer discontinuous. The bordering case to continuous conduction follows with the periodic J,ollll(lary condition II ; tvC
iD(T) = _UA_ RD
113
1 -,,;c1=+==;(=w=;;:T
:===;)""'2 D
[sin(T - arctanwTD) -
e-(T- 0 3) / wTD
[1-
e-(T- 0 3) / wTD]
-!
sin(a3 ;- arctanwTD)]
+ i D(a3) e-(T - 0 3) / wTD,
(8.16)
where the initial value iD(a3) in steady-state is obtained with the help of Eq.
-lo(a2) = iD(a2 + 'Ir) = O
(8.11)
I'rolll Eq. (8.6), UA
) i + (wT
)2 (1 D
+ e-
wTD
7r /
)
sin(a2 - arctanwTD )
+ E (1 -
e- 7r / wTD )
= O. (8.12)
Sol ving for a2, the condition for continuous conduction can be written as
(t
D < a2 = arctan w TD - arcsin [ E.)1 + (w T )2 tanhC W'lrTD )
UA .
1. (8.13)
l<'oJ" (~ach value of the back voltage E a firing angle a2 exists, at which colltinuous conduction begins. With WTD » 1, i.e. assuming a I 1<'1 1'1 'c 1.1 y sllJool.h current, Eq. (8.13) yields the obvious result 1.11<" 1;1.1\1',<: 0["
"2 /fI!
!T
ú;\ cos a -> E
,
wllidl IIW
(8.14) UD must exceed the back voltage
ror
CIlIll.illllOUS load current to flow. > !\ II L!t( ' preceding equations are valid for E < O, that is, for rectifier as wdL as illv(~rter operation. !\ s wa.'; Illcntioned before, continuous current flow occurs if the firing angle is rl ~
'i n (It:,)
= iJ)(üa -+- 'Ir)
lu ' [IC(' \.1[(' Sollll.ipll IIl;ty
> O;
1)(' writl:clI as
(8.15)
(8.15). li the mean value of the current is mainly of interest as is the case with drives, a much simpler solution exists for continuous current in steady-state. By integrating Eq. (8.5) over a half period and inserting the condition (8.15) for periodicity, we find
J
0 3 + 7r
-iD(a3)
=
:;1
1 (UA sin T E) dT RD
=
R1 (uD(a3) - E) , D
(8.17)
0 3
where the average voltage UD is given by Eq. (8.3) . Hence the converter acts as a linear controllable voltage source only if the current in the load branch is continuous. The mean current i D (a2), at which discontinuous operation sets in, Eq. (8.13), is an important design parameter for the possible selection of a smoothing choke, resulting in an adequate value of the normalised time con stant wTD . As is shown in Fig. 8.16 this current also depends on the back volt age E, its peak occurs around E ~ O, where the AC component contained in UD(T) assumes a maximum. With large drives the specification may call for the maximum discontinuous current to be less than 10% of nominal current, i D (a2) < O.lIDr, to limit the ripple current which produces additionallosses in the armature and may impair commutation. However, with medium and smaller drives the smoothing choke is frequently omitted for reasons of cost, so that only the armature inductance is left for filtering the current. This may result in discontinuous current of half nominal value, affecting not only the los ses but also the design of the current controller, as will be explained in Sect. 8.5. Another important effect neglected so far is the finite commutation in terval caused by the impedance in the AC circuit, which consists mainly of the leakage inductance and winding resistance of the transformerj also, line inductances may have been inserted for protective reasons to limit the rate of change of currents (including fault currents) through the thyristors. At medium and larger power rating, the line impedance is predominantly induc tive, R;1 ::<:: O.lwL;\, which for approximate calculations justifies neglecting the f('sisl.iv(~ cOlllponnnt, Tlt(, liuI' :-l ide illdllcLallce L;\, even tholl gh it, is small whcn compared with UI(' illllllri..'I.JlIT I ,,) 'ill UI(' 1011,<1 circllil., cnIlSI'J» \.II C liJlI ~ ('.IIIT( ~ 1l1. to !>( !('. O lll l: a
8.2 Line-commutated Converter in Single-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives
11'1
sLat.e variable subject to continuous change onlYi hence iA(T) and the cur )"('Ilt.S in the thyristor branches can no longer exhibit the discontinuities seen iii Fig. 8.6. Instead, the rise and decay of the thyristor currents and, cor f"(~spondingly, the reversal of the alternating current iA(T) now assume the f"mlll of continuous commutating transients requiring a finite time, called the coJJlmutation interval Tc or overlapi naturally, this effect occurs only with nllltinuous load current because in discontinuous current mode the currents h"gill in each period with zero initial condition. For example it may be assumed that the thyristors 2, 2' of the circuit iII F'ig. 8.15 a are initially conducting and that at T = 0:, Le. with positive lill(, voltage UA, the thyristors 1, l' are fired. Since the currents iD and iA ;\.IC c()lltinuous on account of L D and LA, the thyristors 2, 2' remain also mlld IlcLing at first, so that the four thyristors represent temporarily a short circllit connection, UD = UB = O, decoupling the line current from the load circllit. Hence, from Kirchhoffs laws, it follows with wt = T, = iD ,
(8.18)
= iA
(8.19)
'Í'Fl - 'i T2 LU
LA .
,
diA b + R' A tA
(8.21)
TIl\" illiCial conditions are
(o)
cc.
10'/,::(11') -
- iA (o:)
= iD(O:)
'/';1 -,
"
I,
a+rc
wt =r
Lo I
i
'--~r
!e
i
TI
L A/ RA we find for o: ~ T ~ o:
+ Tc (8.22)
UA
= ~ + (wLAF
i;; i\w p(:ak valuc of the (steady state) short circuit current at the line termi lIalH. 1"1"0111 r'~qs. (8.18), (8.19) the thyristor currents are obtained, (r)
; ' l '~ ( I')
I : ':1 ')
f'in(n)
~ l i r.((I')
I
I 'I
1'--
i
1\
--~,
\
I I I I DI I
I
II I I I I I
wt=r
....
I
wt = r
d Fig. 8.15. Commutation of a single phase converter
+ Tc) = O
(8.23)
after this instant, Eq. (8.5) is again valido The various volt ages and currents, when including the line impedance, are seen in Fig. 8.15 a,b, assuming continuous current iDo All the currents are now continuous functions oftimei in contrast, the volt ages UD(T) and UB(T) exhibit characteristic discontinuities during commutation, when all thyristors are conducting. As a consequence of the finite commutation interval, the line current ex periences a further shift in time resulting in an additional component of reactive power due to commutation, increasing the reactive power caused by the delayed firing control as discussed before, Fig. 8.7 Another effect of the line impedance is a loss in output voltage due to the temporary short circuit,
1tO(T) = 0,
.
1
'
I
/>,
O ai1
c
a
I
I I I I T2111
Ro
iT2(0:
wll(')"(:
;',/"1
Uo
I I
I I
I
.
iA(T) ~ iA sin(T - arctanwTA) -[iD(O:) + iA sin(o: - arctanwTA)] e-(T-a)/wTA
'tA
I
These results apply only as long as the four thyristors are conducting. The commutation terminates at time o: + Tc, when
O,
It.. , 01"':(' of" Lhe llIuch smaller inductance LA, the line current changes rapidly wlll'f"(':ls Llw load current may be assumed to be approximately constant dur 1111', III(' shorL commutation interval, iD(T) ~ iD(O:). Hence, with the abbre
"I:d.ioJl
\
/1
ue
(8.20)
= UA ,
. + RD ZD =-E .
diD . dT
w Lo - -
io'/'I
t1Uo
---,-,,-
+ iT2
'iTi
115
a
~
T
~
o:
+ Tc
i
';'A(T)I ,
iI.
ill( r }l·
c;m;;(';; ii drop
01"
i.Il(~
JIlcan voltage by
(8.24)
8.2 Line-conunutated Converter in Single-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives
II(i
_
"J+
1
L1uD ~
1r
TC
-L1uD (8.25)
UA dr ,
" Wllich may be estimated from the current change in the AC circuito With the mentioned before,
~;illl[llification
RA «WLA, \VI'
lilld from Eq. (8.20)
---
.d"lll)
~
1
-
1r
"J+
TC UA dr
~
1 -wLAiA(r)
I"+TC
1r
"
"
1
= -WLA [iA (a + rc) - iA(a)].
L1uD ~ wLAIDn iD = k iD . UDO .,fiUA IDn I Dn
1 kllce
n
Uoo
(8.28)
The normalised impedance factor k is mainly determined by the leakage reactance of the line transformerj a common value is k ~ 0.05 to 0.10. The commutation interval rc, which usually lasts only a few degrees, de pends also on the firing angle aj this is seen from a rough estimate of the integral in Eq. (8.25)
discontinuous current
"2
(8.27)
which indicates that the line-commutated converter acts like a controlled voltage source having an apparent internal resistance proportional to the reactance of the AC linej of course, this drop of the terminal volt age does not involve real power losses, as it is caused by commutation at the AC side of the converter. As a consequence, the load curves of the converter uD(i D) for 0'= const. are drooping in the continuous current region (Fig. 8.16), even though all ohmic resistances apart from that in the load branch have been neglected. ln contrast to the intermittent current range where the curves are strongly nonlinear, the characteristics in the continuous current region are linear, exhibiting only a slight droop. By normalising Eq. (8.27) with the no-Ioad volt age UDO and nominal current I Dn , we find
(8.26)
the voltage-time-area lost during commutation is related to the flux ('liallgc in the line inductance.
2 -WLA iD, 1r
1r
Uo
~
117
~ 4.J.UD ~ -1
a= 0°
1r
J' .
"+T
c
UA sm r dr
~
M2
. a -v rc UA sm
(8.29)
1r
"
continuous current
which shows that for constant load current the overlap is minimal at a ~, when the AC volt age has its peak amplitude. A more accurate result is obtained by evaluating the integral, Eq. (8.25),
60° 0,5
1\\ ':
'
i
r/ "'it~.
900
a
":n
120°
150°
8.10. Clmractcristics of single phase converter with inductive load
rc
~
arccos cosa - .,fiwLA-] UA iD - a.
[
The knowledge of the overlap is of particular importance when the con verter operates as inverter, i.e. with large firing delay a, because then it must be assured that the commutation is completed and in addition the outgoing thyristors have recovered and are ready to block forward volt age when the line voltage changes signj otherwise an unintentional refiring may occur. Since the back voltage is negative during inverter operation, E < O, this would con nect in series two voltages (E, UA) having (for a half period) the sarne signj the result wOllld be a short circuit condition with a fast current rise which is only limit<~d by the smoothing inductance. This malfunction is called a
Ir (I. f;IIIIÍcÍI')lI,ly 1i~1'I',(' slIIooLhillP; illdllct.:I.Il(,p Ln is employed, s\lell tllnt, tltc , Ippll ' ('llt n' I.II. '('I," I.nilwcI ill i,,( I) lIItty IH' IU'I',I<'d,'d, i ,1 (n ,·1 r,,) ~ 'I:,,((\') :=::!
COllllJllIl.al.ioll hilllr<~.
I (/ ; I' ~q
ifi
(H . ~~n ) 11,1'11 .I' /I'ld,I . ! 1
VI
,II. ii I'," <1m!, pr"lll,d.l."ml ln !.II('
)1I , ~d (' 111"1"1'111.
(8.30)
III )"i,,;. H. I'1 l,lti H di~t1Ll'b<\l\c e is (kr-;cril)( ~ d iu more detail. The invert(~ r '(;:)0", wll<'1l nll I.n 1'!,I 'I'Il Ln illil.in,lIy aI. I.ltl~ r;;IJ(, IÍrl 111', il.ll/ ~l<- n,
1I !IH lIlIl<'d
118
8. Static Converter as a Power Actuator for DC Drives Commutating
voltage
,
Ua
\
I
I ){
c;
, -,
"
\
\
IW
~
mt = "t"
~
mt = "t"
iT1 lJ
' U
!
"' :
T2
I
I~ (
!
Back commutation to 2, 2'
I
c
·-
---'-rf------==',.l...r.oL.,----....
mt = "t"
Fi!;. 8.17. Commutation failure due to late firing of thyristors 1, l'
;ulditiollal firing delay ..:1a for the thyristors 1, l' is introduced. Due to the I"w voJl.agc 7J.A (r) in this interval the commutation proceeds quite slowly as ii. I.al\( ·.s I.illl!: to accumulate the necessary voltage-time-area. The commutat "11', 1,I'iI.lISi(·1I1. is cventually completed at a + rc, when the current through I.lty,i: ;I.ors 2, :l' liaS become zero. There is still some negative blocking volt age 11,/, ~, ' () arross the outgoing thyristors but it is small and the time until the v"II.II.I'.< · 'Ii /I chauges sign may be too short for the thyristors 2, 2' to recover 1.llI' il I'mwarcl blocking capability. ln other words, the time equivalent of the t· .~ I.illrLioll angle, = 7r - (a + rc) no longe r exceeds the recovery time tr which i ~, i IIcr<~;t.scd by the low reversed bias volt age across the outgoing thyristors. Tllc collscquence could be a spontaneous refiring of thyristors 2, 2' and IlI\ck-cOIlIIn\ltation, leading to the series connection of UA < O and E < O I.Ilro'lf~ h thyristors 2, 2' and a subsequent steep rise of the current. The next I'\'/,:' alar Wllunutation of the inverter is possible one period after the attempted fil'illf ~ (Ir thyristors 1, 1', where the controller must now assure a sufficient1y H.d vallC(·d timing of the firing pulses to successfully complete the commutation d, ~spil.<: I.hc irtcreased current. Otherwise a circuit breaker or fuse wo.u ld have 1.0 d,~;t.[' lhe fault. II. is n(~('lI that. th(~ danger of a commutation failure increases with the IlIiq·,I1il.lld(· (I]' lhe c1lL'n~lll and thc line-side ['('ad.ance, since both effeds pro I"" lí 1.1a(' (,Olllllllltatjoll illl.(~rv;d r,: . III ord(~r 1.0 flllly Ill.Ílise nlc. voHagc of lhe lllv,· rl.v r, II ::ldlitl'·1I1. I 11 11 1'. II i I.lldn of t.l1(~ (·xl.lll<'l.i,," 11111',1 0 ,lIlUy h(· a.'is lIl·(·d hy a ilq' III I' I." ",,"11'1>1 ")('1'. 11"w"v"I', iII vil·w "r 1.III ',"I,l i l.I,,,ml '·'''"(1I< ·xity 1111.1 LIli'
8.3 Line-commutated Converter in Three-phase Bridge Connection
119
remaining uncertainty in case of a sudden voltage dip of the AC line voltage (which would unexpectedly increase the overlap), extinction angle control is seldom used except with very high power installations, such as HVDC, where operation of the inverter at the maximum firing angle is important for power factor considerations and the cost of the control system is less sig nificant [F21,77]. The usual practice with drives is to limit the firing angle to "safe" values, for instance a < a max = 150 0 , and to prevent excessive currents by a fast acting controlloop.
8.3 Line-commutated Converter in Three-phase Bridge Connection With the exception of traction drives having a single phase supply the single phase circuit discussed in the preceding section is normally used only for low power applications, such as the supply of field windings. Beyond a few kW there are strong incentives for three-phase converter circuits. The bridge con nection shown in Fig. 8.18 is the circuit most commonly used with thyristors or power transistors; it requires two more thyristors but has a number of important advantages: • The three-phase line is symmetrically loaded in steady state. • The line currents have a lower harmonic contentj as a consequence there is less distortion of the line volt ages than with a single phase circuito • The sarne is true for the direct voltage uD(r) which contains ripple com ponents of higher frequencies and lower amplitudes, thereby permitting a reduction of the filter components and causing lower losses in the load. • The dynamic performance of three-phase converters is superior because thyristors are fired at shorter intervals; hence the reduced delay for control signals permits more rapid controI. ln contrast to the single-phase bridge circuit (Fig. 8.5) where the thyris tors had to be fired in pairs, the commutations in the three-phase circuit (Fig. 8.18) alternate between the upper and lower row of thyristors, so that in steady state six regularly spaced firing transients occur during each line period. Hence the three-phase bridge is a "six-pulse converter" , whereas the single phase bridge with two firing instants per period belongs to the class of "two-pulse converter" circuits. With Hg-arc valves, having a much higher volt age drop than thyristors, center tap circuits without series connection of valves had been in widespread use; they were the natural choice with multi-anode valves with a common Hg cathode. However, since the introduction of thyristors as individual switching dp'vjc< ~s witll sllla.ll physical dimensions and low voltage drop these aspects have h ecllllu' ohsold,· .
I 'lO
8.3 Line-commutated Converter in Three-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives 5
io
iTtt (x-----,.
~r
t
~r
1II.J 1.
-L O
I
I S2 1
-LA i1
!
U1
II RO I
a
!e
rr 1 2
U2
!
5
S1
Uo
iT4
121
13
U3
,,
!
I
b
i
I
::
A!!T1 /1
Fig. 8.18. Three-phase line-commutated converter bridge
~I
i
O::
I
=
I
-----v T1
c
'!'ltc AC voltages in Fig. 8.18 are assumed to represent the line-to-neutral vI)IL:I./.o;(~s of a balanced three-phase system. The thyristors 1, 3, 5 connect the i\ C lillc-I,crminals in cyclic sequence to the upper DC bus SI, the valves 2,4, 1\ LI) !.II(' lower bus S2. In normal steady-state operation and with continuous d l", ·( 1. clllTcnt, each thyristor in the upper and lower row carries the current !'", I:!()"; a]so, the two thyristors connected to one AC line-terminal do not IIClIlllally collduct at the sarne time, as this constitutes a short circuit of the I li: I.I'rlllill a ls. Figures 8.19 and 8.20 show various volt ages and currents in steady-state ('()lIdil.i()llS with the firing angles aI = 45° and a2 = 135° , belonging to r(' r Lili(~r- and inverter-operation respectively. The line inductances LA are 11( 'I',II 'cLcd for the time being so that no overlap between subsequent thyristors oCCIlrs; thc constant direct current iD = ID is again assumed to be generated I,)' :l cl1rrent source. The conducting thyristors are indicated in each 120° illl.<'rval , always comprising one of the upper and one of the lower row; the Jirilll~ aHcrnates between these two groups; hence the output volt age uD(r) C ( 'lI sis1.s of (i0° sections of sequentialline-to-line voltages, for example UI2 = 'If, I '/I. ~ . As IOllg as thyristor 1 is conducting, thyristor 3 sees a negative bias v()lt. agc 1<>1' 'lt2 - 'U I < Oano can only be fircd arte!' 'U 2 - 'UI has become positive. 'I'hi:: "lla(.l1l'a!" ('irillf.!,' instant (r ~. n/O for Lltyrislor 1) serves as a ref (')'( ' 111'(' r()(· 1.11.1' lirill P; ;\.IIJ-\ II ~ ( ~ , which cau ag:l.ill vn.ry fr<1I1l (Y O (lllilxillll1111 (1111.1'11'1. v"It. II,I':c', ,I, ·,·t.lfi(·r) L" rr ( ~""I." (lI d llillllllll 1111 Lp 11 I. VOJt I I,I~( ' 1 illv<'rl,l'r ),
I·
: .
'
27r
wt
='t'
I
:
7r
, 2n
,~3
,
,I
i
d
I I
I
r::-+l
,a 11
3 :IJi~ II n
I.
2nl
wt =
T
Fig. 8.19. Voltages and currents of a 6- pulse converter at ai
= 45°
An appropriate safety m argin 7r - ama:>: is again required which must exceed the maximum overlap plus recovery time tr to prevent commutation failures. With the definitions for the line-to-neutral voltages uI(r)
= V2UA
sinr 1
U2(r) = V2U A sin (r - 2 7r) 3
7r
u 3(r) = V2UA sin (r - 43
)
,
j
and the assnmed simplifications the mean of the output voltage is
8. Static Converter as a Power Actuator for DC Drives
122
8.3 Line-commutated Converter in Three-phase Bridge Connection
dominates in the first, forward blocking voltage (UTl > O) in the second case. This indicates that rectifier operation is insensitive to overload, while at the inverter limit the danger of commutation failure always lurks in the background, because negative voltage is required across the outgoing thyristor for a sufficient time to let it recover its forward blocking capability. ln Figs. 8.19 d and 8.20 d the current ii in one of the AC terminaIs is drawn, the fundamental component of which is again lagging the pertinent line-to-neutral voltage Ui by the firing angle o: (neglecting overlap). A com parison with Fig. 8.6 d shows that the harmonics content of the line currents is reduced; whereas all odd harmonics of line frequency I were present in the single-phase circuit, there are additionally no odd harmonics of frequency 31 in the three-phase circuito Rence the lowest order harmonics of the line currents, assuming perfect symmetry of line volt ages and firing sequence, are the 5th and the 7th in the three-phase circuito 80 far, it had again been assumed that the commutation between the conducting thyristors occurs instantaneously; in reality, it requires finite time because of the presence of leakage inductances in the supply lines and the thyristor branches. As mentioned before, line inductances are often intro duced deliberately to limit the rise of the current during commutation; if no transformer for supplying the converter is used, this is mandatory. Again the line impedance is assumed to be purely inductive, RA « WLA.
3
,I
rr_-
3TC
-",- 4 Inverter
Uo(1:) L)
1/ _.-:
I
-1- - -7 -
,
I
TI ,
wt = 1:
:TC
(;
,II.
r
SI
iT1
5~I
wt = 1:
TS
Uo
I" .r: · )·I.:lO. Voltages and currents of a 6- pulse converter at
0<1
= 135
0
,'l
3 II./i ~ -
/
'Ir
'Ir
i+ a
~+Q
=
3V6 - UA cos ('Ir)li+a r + 'Ir 6 f+a
io
'Ir sin(r+-)dr 6
3V6 =- UA coso:
'Ir
= UDO
i2
coso:.
!
U2
control curve follows a cos-function. When the converter is rl.ed to the 230/400 V three-phase line, the maximum mean output voll,;W;(~ ii-> (1)O 540 V. 'I'IU' olltput voltal~e 'uJ)(r) in Figs. 8.19 and 8.20 contains harmoncis, thc rn'<[I'H'lIfiC':-l (II' wltic.h aJ'(~ lIlultiplcs of G/, i.ç. :\00 Ilz rOl' a 50' R,z Iinc~. V.. II.:q',e' 1\11 <1 ('IIITe'ul. (Ir ii. I.hyriH(.or l U'C' pllll.l\'<1 111 Jo'i, ç. R.H) (' fOI' )·(·('I.ifi er l\cr(' I.on, lhe
' U
1,1
=
I" i". K ~O
e'
1'111"
I lIve'r!.!'I' o\,,'rnU(III .
TI
T5
5
L, Iuo
,----,
j II
LA
11"v" 'I't1(' I tlnl l v "II, I\I ~( '
('11'1'1
O)
a
LA i2
iI
i3
U3!
UI! b
Uo
~.lO
1"8, ,
LA
(8.31) ( ' IIII I ...
fi
i
IT6
f+a
3V6 / (Ul-u2)dr=--UA
_
•
6
6 .. ~+a
io
io
iO
,/ Iii
123
U2!
LA
T2
iI
i3 6
u3!
UI! -
U2!
c
Fig. 8.21. Sequence of commutation in three-phase converter bridge
Th(' COllllIIlIl.al.ion dcscribed in Fig. 8.21 proceeds in three stages: lnitially the tltyriHl.ors ri aliei G lIlay carry the consta.nt current iD = lD (Fig. 8.21 a); nl. I.illll' I ;~ I " 1.I1'y f' isl.or I. is firec] 1I\lI,(Ii)lf~ to ti\(' Ritllation in Fig. 8.21 b,
I~!J
8. Static Converter as a Power Actuator for DC Drives
8.3 Line-commutated Converter in Three-phase Bridge Connection
125
wlwre two AC terminaIs are connected through thyristors 1 and 5; due to the ~:lIlall inductances LA a short circuit current rapidly builds up, transferring U J( ~ current ID from thyristor 5 to thyristor 1. As soon as iT5 has become 'I,n u , thyristor 5 blocks and the circuit finds itself in the position shown in
'f
"'i g. 8.21 c. The overlap during which three thyristors are simultaneously
('olt
+ iT5
= iT6 ~ ID ,
(8.32) ,
i
-:'fc:+
d w L A dT (in - iT5) ~ UI - U3 .
(8.33)
.,._._._.~) ,
~
din
..;' ~._._.J. ,
,
1\ \ I
2wLA ~ ~ UI - U3
;;:.
= V6UA
sm(T
-7r/6),
\
'f
"
i..
(8.34)
,---- -' .-. . \ - iT2
1(':ldÍlIg with the initial condition /'f' I (r)
~
in(~
+ a) = O to
(7r)]
-v'6U - - [ cos a - cos r - 2wL A 6 A
,
'T5
, _____ J,
: o.
1':lilJlillation of iT5 results in
'f
.
T3
- iT4
- iT6
Fig. 8.22. Commutation transients of three-phase converter
(8.35)
-ij-+a+TC
LlUD
:1.11<1
I
3
= :;
~(UI-U3)dr.
(8.39)
-ã"+a /',/" ( / ) ~ J J) i I :,
::0"11
:IS '/''/'I>
'Il '/'/'Io
(
Ij
-
in (r) .
(8.36)
Integrating Eq. (8.34) with the limits O ~ in ~ ID results in
i!) zero,
' I' ü+rc
)
=0,
(8.37)
i+;"'.+TC 1 2' (UI
- U3) dr = w L A I D ;
(8.40)
-ij-+a
ii", ('1I11111l11tatiOll is concluded and Eqs.(8.32), (8.33) are no longe r valido
I"Í/';IJI'(' 8.22 shows a sequence of commutations as obtained by digital :illllld:tI,ioll of the converter circuit [E6, M17] where for the sake of clarity /.lI(' ('OIIllll1ltation intervals have been enlarged. Because of the equal induc 1.;\,\1,,(, :; iii (:ach line terminal, the potential of the upper bus SI corresponds Co LI\(, IIJ(~i).Il of the two line-to-neutral volt ages affected by the short circuit ('olld ÍLÍolI. Hence, the output volt age during overlap is 1
'IIO(T)::::::: ;;(UI +U3) - Uz, whicli l'I~II:Wh n Ins,~ H '~ , 'l ; í 1.,;
i.1l
HI P r" ll \lJ\lIIIIrI. í "l
OllLpllt
(8 .38) vol1:;I,f2:(~ accor
the voltage drop caused by the commutation is again a linear function of the load current,
--
3
LlUD = -wLAID;
7r
a fictitious ohmic resistor in would have a similar effect. verter in Fig. 8.23 show a range. As was cxpbilH:tl in the
(8.41) series with the controllable volt age source uD(a) The load characteristics of the three-phase con corresponding droop in the continuous current
case of t:hp sillgJe-phase converter, a load CÍrcllit illdllctivc: illlpcdaw',· iII : :<'ri(~s with a back voltag( ~ ]i) Citll 1<-0<1 1.<1 di n..c"t!,'IIIII
consistill/,: "r lt
(illil,,'
8. Static Converter as a Power Actuator for DC Drives
126
Ilighly nonlinear; this is seen at the left of Fig. 8.23. With an appropriate clloice of the smoothing inductance LD the region of discontinuous current C
uoo discontinuous current
~:1~; : /: 1
a=O°
\ \
~inuous current I
~
\
\
I
-
a= 15 0 io
,.'
90 0
r
/
"'1\ ( H.4:1. I,oa d
The dynamic behaviour of the six-pulse converter is similar t0 that of the two-pulse converter with the exception that now the steady-state firing intervals are 1r/3, instead of 1r, permitting faster access to the output volt age UD(T). The waiting interval is now between O and 1/6 f corresponding to 3.3 ms for a 50 Hz-supply. The firing control of a three-phase converter is simply an extension of the firing scheme for single phase converters (Fig. 8.11), employing several time-displaced saw-tooth functions to sample the output signal of the con troller. For example, Fig. 8.24 displays voltage transients UD(T), neglecting commutation, which follow a step change of the firing angle from rectifier- to inverter operation and back. When the firing is delayed, the current carrying thyristors simply remaín conducting until the line volt age has changed its sign, while in opposite direction an immediate response is possible. It may even occur, as seen in Fig. 8.24, that some thyristors are skipped because of a multi pie commutation with several- in principIe ali - thyristors temporarily conducting, with the ones connected to the highest and lowest line volt age potential eventually prevailing.
0,5
tI /
127
60 0
\ 1
8.3 Line-commutated Converter in Three-phase Bridge Connection
a
150
IOn
n+,g. --::c;;}.,\
:1 ":2,
multiple commutation
n
2n+ "6
wt=1'
0
curves of a 6- pulse converter
'I 'Ice l.clllporary short circuits of the AC terminaIs during commutation lIavc ()[ (".OUfse an immediate effect on the waveforms of the AC voltages, pos ,:ildy inlluencing neighbouring equipment. One of the terminal voltages UBl wil.h rc ~SP( ~ct to neutral, Fig. 8.18, is plotted in Fig. 8.22. ln a weak supply sys 1.('111 haviug a large internal impedance, the danger of inadvertent feedback to 1.11(' ("'Ollvcrter through the synchronising signals exists which may even cause ill.~ l.a.bility [A11]. Apart from the obvious, but not always practical, solution (Ir lill.c rillg t.he synchronising signals, there are various "supply-independant" lirilq( S(".hClll(~S wit.h internal oscillators that permit stable operation even with hil':hly di:;l,oJ"l.cd lill( ~ volt ages [H61,M16]. Three-phase firing circuits are abo nVl~ih"k iII 1.11< ' (onll ()r il)I.(~gral,cd circuits (chips). .
Fig. 8.24. Control transíents of a 6- pulse converter
A feature of the three-phase bridge connection is that two thyristors have to be fired simultaneously for start-up and in discontinuous current mode, one in each rOWj otherwise the current could not begin to flow. This is achieved either by employing wide pulses (LlT > 1r/3) so that they overlap or by generating for each thyristor two short firing pulses, 1r /3 apart. Normally the second pulse would be applied to a thyristor that is already conducting and be of no consequence. If the first solution is preferred, the firing pulse may be chopped into narrow pulses to redu ce the voltage-time-area and the size of the pulse transformers. The iupllt current ofthe converter, drawn in Figs. 8.19 d and 8.19 d with oul t;ü.illg COllllllllt.a.tioH into account, is the sarne as the current that would fl ow iII I.hl' prilll1\.ry willdiug of a Y /Y- <:ollll(~ded transformer supplyill g the convI'rl." I". Illlw" v"r, til.· w:l.vc{llnll nl" ti II' prill llu''y 1·.lIrJ"(~1I1, dlall f1:( ~ S ir (III( : 0(' I.h(~
128
8.3 Line-commutated Converter in Three-phase Bridge Connection
8. Static Converter as a Power Actuator for DC Drives
129
2ia
ia
io
J'O
i3
~'
n' ~
: 2nl
r
a1
iO
a1
ia
b
U
U
o
a2
a
a2
III
U
02
o
~ b
Fig. 8.26. 12-pulse-converter consisting of two 6-pulse converters in series- and paraliei connection
ip2
i p1
ip3
a Supp/y
F ig . 8.25. Three-phase converter supplied by Y /6- connected transformer. (n) C ircuitj (b) Waveform of primary current
I.r,wsJ"ormer windings are .6-connected as is explained in Fig. 8.25. Assume 1.";tI, Lhe constant load current I D is initially flowing through the thyristors ·1 a.lI
1' )1 .~alce of simplicity - equal numbers of turns, the following equations then Ilold for the primary currents Zp l
=
i3 -
iI
= 2iD ,
i p2
= iI
-
iz
= -iD
'lp3
= iz -
i3
= -iD
(8.42)
,
Pecrllutation of the six possible states yields the primary current waveform /1'1 (T) shown in Fig, 8.25 b which also contains the 5th and the 7th as the lowesL ord(~r harmonics. With the appropriate ehoj('(õ of tlw llllmber of turns !.II<' Y -
1·. IlLTI ~ lI!. wavI~f()l.'lll S .
1'\n ~qll'~ltI.l'y in I.h(~ Jla.~!. ,
o!.l.ailI;dti"
wi!.h ii, ('OllVI ' r!.l ' r
II" I.h .' 11. I.h ,y l'l n!.ol'lI " .lI d Lo
I... JlllÍ liI'd iI'
1)('
Lll
t'llllllIlOld ,l' 1I0W, (,III! oulput volLal1;t~
11IWillU () II(' I.h'yl' in l\ lI ' 1"' 1' l'I'H. llch WHH ill s llfl ir in ll!. (·(lIllll'(l.I'd i II
nn 1'1 I'r! . ~i (l 1\ 1I' 1 ~c1vlI, itl , HW ' R !lla'y !.lil'lI
1111/11·,, <1 III' IlO' ril ':: (,olllll'l '111'l'. l w ll vl dl d d 1," Y l l il l.ol'/i
(·<)llIpll'i.l'
three-phase circuits are connected in series, as seen in Fig. 8.26 a. Rather than taking precautions for equal voltage sharing between series connected blocking thyristors one now relies on the output voltages of the transformer to produce the correct voltages. AIso, if a Y / .6-connection with its inhe1' ent 30° phase shift is chosen for the secondaries there is an added benefit in the reduction of harmonics on the load-side, where only multiples of 12 f occur, as weU as at the line-side, where the harmonics are those of a 12-pulsc converter, with the 11 th and the 13 th being the lowest harmonics; there is a further improvement in dynamic response because the maximum waiting interval is now 30°. Another similar circuit is seen in Fig. 8.26 b where two six pulse converters supplied from a Y / .6-connected transformer are connected in paraUel, employing two separate chokes or an interphase transformer for summation of the currents. The principIe of achieving 12-pulse operation by se1'ies connection of complete converters is applied also on HVDC transmis sions with hundreds of MW, where many thyristors are still series connect(·~ tl in each converter branch. This brief introduction to the fundamentaIs of line-commutated converl~ ers seems adequate for dealing with most drive problems. A more detailcd analysis such as harmonics, circuitry 01' protective measures is given in tlw specialised Iiterature [50, 58J. It should have become clear that multiphas(~ static power converters are electrical actuators of superb dynamic perfor mance providing an excellent tool for controlling large output power (MW) with low powcr control signals (mW) in ve1'y short time (ms). This combjna~ tiOIl is \I11lllal.dl\~d hy ;),IlY othe1' type of power actllator. On the o the!' hawl iI. :;\Jollld 1)(' )lol.,'d I.Ilal. dlw 1.0 I.\w switchilll-!; adioll sL;t.tic
pow( ~ r COII\lI'I'I., ' I':; arc'
1:-)0
8. Static Converter as a Power Actuator for DC Drives
inherently nonlinear. Since they have no internal energy storage the output power is instantaneously transferred into the AC line causing harmonics and reactive power in the supply system; strong grids are therefore required to support the operation of large power converters. Even shorter response times are feasible with force commutated converter circuits, discussed in Chapo 11; while the power ratings available today are adequate for drives, they do not yet reach the leveIs needed in energy supply systems.
8.4 Line-commutated Converters with Reduced Reactive Power I II the preceding sections it was shown that line-commutated converters draw l"<'active power from the line; this is inherent in the principIe of control by ddayed firing which results in lagging current at the AC side of the converter. Siuce the phase shift of the fundamental component of the line current is dircctly related to the firing angle a, the situation is particularly sever e at n ~ ~, where the line current is, apart from the losses, purely reactive. This ('olJdition is unavoidable, if the converter is supplying the armature of a DC IlIotor because when the motor starts from standstill the induced volt age is illitially zero and the converter must be controlled dose to a = ~; hence
-
iO
2'
iD(T)
~
O,
UD(T)
~
O
restrict the operation to a single quadrant in the UD, iD-plane; with traction drives employing series wound motors this would not be objectionable.
iA
-
uA i02
DI
O2
i
":f'\ í\ I.
io
iOI
TI
i T2
TI
T
Ô
a
n a+n I
2
a
O I iOI
wt == T
.
I
I
I
I
. I
I
I .
i
02
I I
wt == T
-r=r::-=t-f----~.
1'
O
b Fif.;. 8 .28. flalf-controlled single-phase converter. (n) CirclIit, clial'.ralllj (h) Waveforms of volt.rtgps nao currents
'1'111' Hil.lllll.ioll 11I1I.'y I)(~ t.oh~rablc wil.h l:i t.r:onf~ s lIpply sy s t(~ IlI S huI. is Illlac ""pl.ld,I,' "II 1. J'lII'Ll lIlI driv" ll, wl)(')"(' t,1i, : r·(':t('(,iv(' ('li rn''lIt, iII c(lluhillaLioll wil.h
I I I I
---CTt-_:Tlj-t I
,
8.27. Single-phase converter with a shunting diode
~+2n wt==T
I
Uo
I"i~~.
131
the large inductance of the power supply comprising the catenary would cause heavy voltage fluctuations, in addition to the harmonic distortion. Mainly for the benefit of traction drives special circuits and control techniques have been developed with the aim of limiting the reactive power at reduced output volt age. Some of these circuits contain diodes as well as thyristors [D25, S57]. A simple though uneconomical solution is the use of a diode parallel to the load, as shown in Fig. 8.27 for a single-phase bridge circuit. The diode has the effect of cutting off the negative edges of the output voltage, essentially acting as bypass during the intervals when UD would otherwise be negative. Thus the trailing end of the line current (during interval1l' < T < 1I'+a in Fig. 8.6 d does not flow through the converter and the line as it is short circuited through the diode. The result is an offset of the fundamental component of the line current in leading direction and a correspondent improvement of the power factor. This has, of course, the effect that the circuit can no longer act as inverter; the conditions
a,1
UA
iII
8.4 Line-commutated Converters with Reduced Reactive Power
wt == T
8. Static Converter as a Power Actuator for DC Drives
1:\2
8.5 Control Loop Containing an Electronic Power Converter
Another circuit with similar properties is shown in Fig. 8.28 a, which I,dongs to a class of half-controlled converter circuits containing diodes in "lace of some of the thyristors. Here too, the line current cannot flow against I.I\( ~ line voltage and thus cause a temporary power flow from the DC to the I\C side, since the opposite diode would conduct and act as bypass. Figure K.2 8 b shows a simplified diagram of the volt ages and currents, neglecting IIw commutation intervals. Again the trailing end of the line current is cut ,,:r, tltereby advancing the fundamental component and improving the power f";wto!". Clearly, this circuit could be easily extended to form a half-controlled 1I1r< "'-phase bridge connection. Uo
iO 2U01O
a=O
8.5 ControI Loop Containing an EIectronic Power Converter
2 a= _____ _ _______1f. UOIO '
iO "' i l~ '
1' . 2 ,. . Bonsl.·· and -buck connection of controlled and uncontrolled converter
i\ 1.,·('I'lIiqll(, ror improving the power factor at the input of the converter ,:: 1)111.11111 '" iII I"ig. 8. 29 a, showing a "boost and buck"-scheme of a three 1,1,11: :" I< '("t ili" r co llt.aining only diodes, and a three phase thyristor converter; 1,\, "IIIJll oy illg a transformer with Y / 6. secondary windings the harmonics :,il.lliI.l.ioll Ill ay oe improved as well. This connection makes use of the fact 1. 11:1.1, ;, CIlllverlcr operating at a = O or a = a max has a tolerable power factor wltil<- ojlcral.ioll at a ~ ~ should be avoided, if ever possible. The total mean IlIII.plll. voltage, assuming continuous current, covers the range II"
Ifi
+ Un 40 cos a m
1\X
:S
UD
:S UD
10
+ UD 20
.
(8.43)
H y .';( d('("I.illg thc voltagcs according to the condition I I , ) lO I (lo ~IJ (:OS Ü'''" X , ~C
in the centre, U D ~ UDlO, where the controlled converter, accounting for about half the total power, is fired at a ~ ~. At full output voltage, a = O, the circuit corresponds to Fig. 8.26 a, causing approximately 12-pulse line interactions. The principIe of boost-and-buck operation may be generalised by splitting the full voltage range ±Uo and assigning the bands of ±Uo/n to n series connected thyristor converters which are supplied from separate transformer secondaries. The control is then arranged on a sequential basis such that only one section is phase-controlled with O < a < a-max, while ali n - 1 other sections are operating either with a = O or with a = a max · 8equential phase control is of particular interest if the converter needs to be subdivided anyway for reasons ofthe high voltage required. This technique has also found application for rai! traction drives [K70].
-~--~-- - _ . _~---
(1
Uo
133
O,
1.1,.. ( ' Illllpld( ~ lir ~: 1. (jlladralll. of I,\w 'iTiJ, i o -plnw' is within reaeh. Tha.t. this cl l ('llil. 1i:1 :1 a 1)( ' 1.1.1' 1' l"( 'aC"i'.iv(' »OW ('[ hablill<"l' til /uI. a sirll~I(' c()n :v( ~ rt.( ~ r cin:lI;it r" II I1 WII II'IHII 1,11<" elido I.h a l. rOI" JIlillillllllll IUI" !II H.lriullI!1I voll.n'I':<', i.l' . "11, /1 N () IUe! " I' "I ' ,,, ,, ,, h,, 1. 1t ;i. ·, ' Id "lI lI "I ",,'td ,,' wi U, /',1" ,, 1 I'0 W"I ' l'ilC"i,or. II. iii ,,"I y
Line-commutated converters are ideal actuators for electrical drives; a main feature , apart from their simplicity, proven reliability and practically unlim ited output power is their excellent dynamic performance. AIso, electronic controllers can be directly connected to the firing circuit which consists of integrated circuitry and logic devi ces operating at signallevel. When a controlloop containing a converter is to be designed, the question arises, how the dynamic behaviour of the converter should be described by a mathematical model. The difficulties stern from the fact that the firing of a thyristor is a discrete process - the firing angle is not a continuous function of time - and that the steady-state and dynamic behaviour of the converter is highly nonlinear. Pulse- and phase-modulated systems of this or similar kinds have been frequently dealt with in the literature but no coherent theory exists which is simple enough for general application [G31, L3, P46, 824]. A detailed analysis of dynamic processes in line-commutated converters is rather involved; the transients are described by nonlinear difference equa tions which defy an analytical solution. Linearisation of the equations is of course possible but the validity is then restricted to the vicinity of the chosen operating point. However, as a first order approximation it is found that the converter dynamics can be entirely neglected as long as the control plant, fed by the converter, has a low-pass characteristic, the controller contains an illt.egrating term and the controlloop is to be well damped [F12, L35]; these couditiolls are usually met on drive applications. Near the stability limit this Silllplificatioll is of course no longer valid; there it is necessary to revert to Lll( ; ac(".lIril. t. ( ~ IlIodd ( ~ qllations . 1\11,,1.1[("1" :: illlpk ll\( )( ld tha t. lms serv( ~d wdl ill practice is based on the oh:;(' l" v; Ii,j"lI 1.11:11. ; 1 varia.hl!'. w;I,it.illl ~ 1)('I"i(1(1 "!:q)s('s III1t.il a s lIl a Il cha.ng(~ of t,he
1:J1
8. Static Converter as a Power Actuator for DC Drives
8.5 Control Loop Containing an Electronic Power Converter
illput signal Yl to the firing circuits (Fig. 8.11) shows an effect by shifting I.lte first firing pulse. As the waiting period depends on the instant at which I.he input signal is changed, varying between zero and 20/6 ms = 3.3 ms for i i si x-pulse converter operating on the 50 Hz supply, a mean equivalent delay I.ime Tr = 1.67 ms is postulated [52]. This heuristic mo dei produces useful n ~ slllt s , as long as well damped transients of the closed loop are the objective IIIl t étgain, as the stability limit is approached, considerable deviations from 1.1)(' predictions must be expected. The most effective method of neutralising the nonlinear and discontinuous 1l('!J
e Uoo
io Ref lon
---
II)
nuf
~
,
io
e!
--",
~=--
io
Fig. 8.31. Simplified block diagram of current control loop
the current sensor is installed on the AC or DC si de of the converter. Sensing the line current is simpler because current transformers with an output recti fier offer inexpensive isolation; a DC current sensor in the load circuit, which may be faster, more accurate and responding to the sign of iD, is likely to be more costly. The use oflow levei control signals, typically 10 V, which are electrically isolated from line potential is a necessary condition for safety rea sons. When designing the controller, the delay of the converter and the filter time constant, which are both small in comparison with the time constant T D , are lumped together to Eorm an equivalent lag Te ~ 4 ms. A proven design method for the current control loop was discussed in Sect. 7.2. lnstead of the PlD-controller that would be desirable with a slowly responding rotating generator, a Pl-controller is quite adequate here. ln fact, a PlD-controller, whose step response begins with an impulse, may cause inconsistent results in conjunction with the discrete firing circuit so that the additional lead term is usually omitted. Hence the open loop transfer function oE the current control scheme is
~Gc,i
T i s+l 1 GD TiS T e s+lTD s+l'
(8.44)
where Gc,i is the gain and Ti the integrating time constant oE the current controller. A typical time constant of the load circuit is TD = 30 ms. Because of Te « T D the transfer function may bc simplified in the region of the cross-over frequency, which determines the band width of the closed loop, FoL(s) ~ G c,iG D rn
1 rn
T i s+l
-~ Te S + 1 .
(8.45)
L I_ _ _ _ _ _ _ _
Fil{. 1-1.:10. COllverter with current control
'1'11 (' circni(. in Fi g. 8.30 is now transformed into the linearised block dia iu Fil';. R.:n , wh( ~ r< : the converter with its control circuit is approxirnatcd I,.v 11.11 (:qUiVldl 'lll. 111('11.11 dd<J,y, which for a fi-pul se umvcrter and 50 Ih-sllpply II 'l ~" r.~ 1.'1 111.' 1 I,'m ::IIl()(ll.hiug t.iI,' I·UlT('1I1. rl 'I'llbac/( sigual) TI:' 2 IIlIl IIlay III ' 111 11 11 11111',[ wil.h Il In x 1'111 11" 1·IlIIV,'rl."I' . 'l'lli:! 1111 111<' .1"1)"11.1 11 all'l ll "II whl'l,IIl 'I'
1 ~ 1 1\l1I
}Q.. lon
FoL(s)
Lo
135
This is in line with the "symmetrical optimum" mentioned in Sect. 7.2. The best choice of parameters is [KI8,37]
Ti wltl ~n '
= a2T~ ) (f.
GOL
= Gc,iGD = ~a TD Te
I I "2 J) f'ollowti
Jlai r (lI' l'I I\" lIV: t.lIII':';
II
frOI/l
I 01' f)
'
(8.46)
t.he Iksin'd dalllpillg ratio D of the complex 1/ ) 2 ir; fr<'qlll'lIl.1y ('hos(:11.
I :I(i
8. Static Converter as a Power Actuator for DC Drives
The design of the current controller leads to the closed loop transients sllOwn in Fig. 8.32. The curves were obtained by accurately simulating the opcration of the three-phase bridge circuit, including the leakage inductances (,1' Lhe transformer and the thyristor branches. The effects of the commuta Liotl are seen on both AC and DC voltages uBdT), UD(T). The transients illdllde step changes of the current reference as well as large synthetic load .I isl.lll'bances caused by inverting the back voltage e. Ikspite the severe simplifications made in designing the current controller, t.11I' I,l'
WVV\f1A
f\f\f\A 9 til \[Jll
I+-+l 10ms
I ~~
137
8.5 ControI Loop Containing an EIectronic Power Converter
The transients in Fig. 8.32 are better damped than would be expected on the basis of the assumed parameters (a = 2, D = 1/2). The most likely reason is the fact that the assumed mean delay represents a somewhat pessimistic approximation of the dynamic response of the converter; another cause is the sim plification 1 ----~
TD
S
+1 ~
1 -- TD S
'
when applying the rules of the symmetrical optimum. The operating conditions assumed for Fig. 8.32 were so selected that con tinuous current prevails except for one brief interval, where the load current becomes discontinuous. As mentioned before, the current is in general not al ways continuous; for example, if the smoothing reactor is omitted, the region of discontinuous current may cover a large portion of the operating range. If a current controlloop, the controller of which has been designed on the basis of continuous current, is operated in the region of discontinuous current, the transients become very sluggish and the quality of control deteriorates. The reason for this is that the gain of the converter is greatly reduced when the current becomes discontinuous; in addition, the load circuit is no longer characterised by the relatively large time lag TD, as the current begins at zero in every firing interval, i.e. every 3.3 ms with a 6-pulse converter. Therefore the PI-controller, Eq. (8.44), is poorly tuned to the controlled planto If consistency of the dynamic response is important, which is the case for instance with servo drives on machine tools, the current controller should be readjusted during discontinuous current operation, i.e. use of an adaptive current controller becomes desirable [B90, 835, 848]. The fact that the volt age gain of the converter is reduced with discontin uous current is qualitatively explained with the help of Fig. 8,33 for a 2-pulse converter .
.~ , ________JI~t1~__~~____~____~~__~~~~~~\,::oé~.~ Discontinuous current I,'i,<, H.:.12. Computed transients of current controI Ioop
Wlicll studying the details of Fig. 8.32 more closely, differences become at time tI, t2, following the inversion of the back voltage e. The ,( ':J,;;O Il is t.hat at 1;1 the back voltage was reversed immediately afier a thyristor ",1..1 hCCll fired, while at t 2 the reversal took place slightly before a firing i'1I1 ~; (' was dll('. Rellte a waiting interval was unavoidable at tI, whereas at I:~ 1./11' ('(lI1v('r[,(~r could rcspoJld immcdiately. The differellce in overshoot of LI\(' \"IITCIII. is i\. COIIK<'ljuc nce o[ Lhe discOlLtil1l1 o u~ nud lloulillear n a.tllre of tlw (·/ 'II V.TI.pr. Tbt' l'onlr,rll l lNolllly JHI,r tially Sllc.c('s,qfll.l iII ('.OlllpcllsaLÍ.JII'; L!tis dl(~d" 1'·( ' n~II: ;( ' 1,1,11' d l ll!.u rlI HIJ( ·(' i i'; 1I.(' 1.i1l,~ ,lil'('d.I:,Y 1111. 1./11' plalll..
wt =-r
ai'l'ar(~llL
Pig. 8 .33. VoItage gain of converter for continuous and discontinuous current
If wil.h cO ll lillUOIIS t \lrrcnt (ú :::: ú:J) the Jiring angle is reduced by a small 1.1", p() :l il. iv( ~ V()ltlLg<~·· l.iTll(~ ;tn'lt, n ,.. "r < 7f, is illC1'ci\sed, whik tlw
11.11 Itlll IIt. /\(~,
1«'I',nl.i",' v"IL IIf'," U III<'
1\,1'('11" /I
•
T'
/I
I
11' ,
i: i I'('dll('('(\' Wit,h disc()J1I . iIJII()II :~
1:\H
R. Static Converter as a Power Actuator for DC Drives
9. Control of Converter-supplied DC Drives
iI! II
fi
1\ fi
fi
fifi II
f\ ~
10ms
vv
Fil!:. 8.34. Step response of current controlloop with adaptive current controller
CIlITCnt (o: = o:d the situation is differentj the interval, where UD(T) > O, is st.ill illcreasing when o: is reduced by Llo:, but so is the time, where UD(T) < O. IÍPJl c:e, the voltage gain dUD / do: is much smaller than in the case of continuous ClllTent. The block diagram of the current control loop for discontinuous current colTesponds to that shown in Fig. 8.31 but the normalised voltage gain of the (·()Jlvcrt.er is now considerably lowerj also, the lag element with time constant ' /'i> is llOW replaced by a lag having a time constant approximately equal to \.II<' liliul-\ interval T D ~ 3.3 ms, which is about l/lO of the former value. ( :I( 'arl y t.his calls for a different design of the controllerj for example, an illl.( 'I',I:\1 colltroller with the transfer function , I,' ( . . ) .', ' .
1 7."i S
with
Ti =
G'o (Te + T1)
,
(8.47)
,ILII :,('l.l'Ii s(,,1 hy a small time constant TI, i.e. high gain, could now be ex 1>1'1 1."" 1.,) pr()duce fast and well damped closed loop transients. III md
Static converters are ideal electronic actuators for DC drives because of their practically unlimited output power and excellent controllability. The speed of response is usually adequate to handle the electromechanical transients occurring in drives. Line-commutated, phase controlled converters or, as they are also called, converters with natural commutation, are the most frequent choice for industrial applications, where a three-phase supply is availablej this is due to the simplicity of the circuits requiring a minimum number of active and passive components. For vehicle drives, where no AC catenary or independent AC supply is available or where the reactive current and the harmonics caused by a line commutated converter would be unacceptable, it may be necessary to employ forced commutated converters having a more complex circuitry and involving higher lossesj a special situation exists also with small DC servo drives, where the response of a line commutated converter may be insuflicient to cope with the stringent dynamic demands and where a chopper converter supplied by a DC link and operated with a higher switching frequency is necessary.
9.1 DC Drive with Line-commutated Converter When connecting a line-commutated converter, for example the three-phase bridge circuit in Fig. 8.18, to the armature of a DC machine (Fig. 7.2), it should be kept in mind that the converter can only operate in two quadrants of the uD/iD-plane as seen in Fig. 8.23. Because of UD = Ua , iD = ia it follows from Eqs. (5.6, 5.8) that, assuming constant sign of the flux
9.1 DC Drive with Line-commutated Converter
9. Control of Converter-supplied DC Drives
1'10
Braking of the drive in the 2nd quadrant by regeneration is not imme di a.Ldy feasible; of course, there is always the possibility of non-regenerative 1)J;tking with the help of external resistors. Two-quadrant drives of this type are suitable for unidirectional loads, Lhe t.orque of which contains a large component of friction, such as paper- or I'rillting machines, calenders, also for pumps or fans; in principIe, it would also IIC applicable to hoists without self-Iocking gears, but rever se torque is usually (I('('ded there too because of counter weights or for helping to accelerate in LlJ( ~ lowering direction. With the driving torque being proportional to the product of armature ('lIrrent ia and flux Pe there are two options for achieving reverse torque, i.e. f()lIr-qu adrant operation: reversing the armature current or the main fluxo 11()(.b lIlethods are in use, though with markedly different preference. 1
2
3
La
iaRef (11/ 1,)1
ia
(I)
~ ia
before the switching command is actually issued; the firing pulses must be inhibited until the commutation is completed and the outgoing thyristors are blocked. If the current transducer senses the load current at the converter side of the reversing switch (as is shown in Fig. 9.3) or with current transformers at the AC side, the current controlloop is unaffected by the switching state. However, the sign in the speed controlloop is reversed due to the fact that a given output current of the converter now produces reverse torque; to offset this effect, one more inversion is needed at the input of the speed controller, as is also indicated in Fig. 9.3. Another possibility for reversing the torque of a DC motor is to reverse the flux Pe while leaving the direction of the armature current unchanged. This can be done by reversing the field through contactors or a bidirectional converter (of considerably reduced power than for the armature). However, since considerable energy is stored in the magnetic field, flux reversal requires llluch more time than inverting the armature; even when applying 3 times Hominal field voltage, the reversing transient may take up to a second on large drives, during which time the armature current must be kept at zero; hence, a lI oticeable zero-torque interval occurs, which is detrimental for the response of I.he controi. A motor subjected to this duty must be fully laminated; also, the voltage induced by transformer action in the armature must be considered. For these reasons field reversing is rarely applied today. There is now a general preference for fully electronic reversal of the ar Ill ature current with the help of bidirectional or dual converters. The cost of I.ltcse schemes has been reduced as a result of advanced semiconductors and IlIanufacturing techniques. The most common circuit of this type is shown in I"i g. 9.4; it consists of two six-pulse bridge-converters C 1 and C2 , connected iii opposition. Clearly, to exclude short circuits between line terminals, only
II·IH . 11 .1. (:;lHc ad e control scheme of DC motor with two-quadrant converter úJ
'I'IJ(~
ó\.I'I11aturc can be reversed with the help of a reversing contactor or, as iII I''ig. 9.3, four thyristors, of which one diagonal pair is conducting. .'·;ill('(· I.lIyristors cannot be blocked electronically, armature reversal can take I'lan: ()uIy at zero armature current; hence, a switching command is only .. lr(·(·Live WIWIl the conditions
úJ
11.,'(
= O,
iD = O
~ i llllll(.atll·(Hlsly
(9.1)
fulfilled. The first signal indicatcs that the superimposed :lP('(',[ ("oul.rolleJ cnlls for a r<~v<: rsal of tOJ:quc while I.Iw scconcl confirms that LI)(' ( ' 111'''''111. COII (.r()1 I('r bas n :dllccd Lhe artlla(.IIJ'(' 1'\11' 1'('11(. (.0 'I.ero, p oss ibly after 1.11<' (·OIIV( ·I'(.I'l' hU H 111'( ' 11 opnal,iul,( (.(,lllpOJ'II.J'il y l1 [il l.l1 ill v<,J'I.('I'. SillC(~ I.lwn! lIlay 1)(' 1;<1 1111 ' 1111I ·PI·I.III II1.'y wil." !.II(' ~: ('('(llId ('(111<1 11, 1,, 11 I II I.I!, . rlU' l' 01' i II I... n II i 1.1.('11 I. 1'1111"111. , ii . 1/1 11111'(1,1. .1111. 1.,. \V .lil. Ii :.11<111 1111 111Vii i \''1 /1.1> 1.1", '·(lII V('II."1 1;1 .. ..rwd,
111'( '
1
=úJRef 1
Field weakening
----------\~
::l!oWII
'i. "
141
1 O
úJ
" -... úJ= úJRef2
Eftect of current limit
7
,./
Unidirectional I:urrent of (:unveI1er in ;> O
I" l l~' 1I,:l.
ia = ia max
Speed control
mn
2mn
úJ= úJRef3
()I' ''I 11I.i ll/ \ rltl I!',"
"r a
I.w" ··<[IIi1.dr;',1I1. dr.'l v"
m
142
9. Control of Converter-supplied DC Drives
9.1 DC Drive with Line-commutated Converter
Supp/y 123
a torque reversal, for instance ia Ref < O and the current controller responds by lowering the armature volt age UD = U a with the consequence that the armature current eventually reaches zero, ia = O. As soon as this condition holds, the firing pulses for the conducting converter CI are blocked by the command module, while those of the incoming converter C2 are activated. Hence the switching conditions are
úJRef úJ
-
~
;0 ;ORef
143
[ia Ref > O1\ ia = O] --+ CI On, [iaRef < O1\ ia = O] --+ C I Off,
I. I
;0 úJ
Fig. 9.3. Four quadrant operation of a DC drive by armature reversal
one of the converters can be allowed to conduct at any one time; hence only one thyristor of each pair produces conduction- and switching- losses so that the pair can be mounted on the sarne heat sink. However, having opposite polarity, they must be electrically insulated. At lower power ratings, com plete thyristor modules are available, having the necessary interconnections built-in. Supp/y oncommon heat s;nk
(ú
ia
úJ
Fig. 9.4. DC drive employing a three-phase dual converter without circulating current
Tlw cllrrent controller supplies
C2 0ff, C2 0n.
Additional features must be incorporated to enhance the safety of operation and to improve the control transients. For example, a waiting interval of 2 to 5 ms may be included before firing pulses are issued to the incoming converter, to be certain that the current in the outgoing converter is really zero. Activating the firing pulses of the incoming converter before alI opposing thyristors are safely blocked would cause a line-side short circuit current that cannot be suppressed by control but must be cleared by fuse links or by a breaker. On the other hand, the firing pulses for the outgoing converter must not be blocked as long as current is flowing because, should the converter operate as inverter, this could lead to a commutation failure, as explained in Fig. 8.17. Therefore, the decision to switch from one converter to the other is of considerable consequence; in particular, the state ia == O must be ascertained safely and with a high resolution. For instance, a converter having a nominal current of 1000 A may contain two arms where at the criticaI time a residual current of 100 mA is flowing; if now the opposite converter is turned on, a short circuit is still likely. Sometimes two current sensors are employed on large drives, one having a linear characteristic over the full current range, to be used for feedback control, while the other serves as a zero-current detector for the command module; its characteristic would have to be linear in the low current region only. AIso note that the currents through the snubber circuits may affect the current measurements, increasing the noise of the signal. Clearly, great emphasis must be given to the reliability of the sensors and control components if this circuit is to operate safely under practical conditions. lndeed, this has initially been a problem that has delayed the general use of this exceedingly simple converter circuito ln the meanwhile I,ite problems have been overcome and reversible converters with antiparallel I.hyristors are produced in large numbers in the form of very compact units covering a wide power range from a few kW to about 10 MW. For the higher ratings modular designs with air- or water-cooled heat sinks are available, SecL ll.4. Hcsides the control scheme shown in Fig. 9.4 there are a number of vari ;d.ioJls; for example, only one firing circuit might be used with the output Plll:;(':; heiJlI'; switdu:d from OIlC converter to tlte other. AIso, the drive is usu idly (·xl.(·lId(·d Lo illdlld(: lidd w('alwlliul': Il~; diHClI: ,:;"d ill S(:c!:. 7.3. Thc field
9. Control of Converter-supplied DC Drives
1~~
wíllding would then be supplied from a separate unidirectional converter of :;IIl;dler size. Thcre are other ways to further improve the performance of the control. '1'111: adaptive current controller has already been mentioned in Sect. 8.5; also, 1.11(' ('(lrrent controller may be given a suitable initial condition, to optimise 1.11<: cllrrent transient after the zero current interval, when the incoming con V ( ' rI .( ~ r is activated. The best initial condition depends on the existing current rd('n ~ IIc(' and the magnitude of the back volt age which is known from the lirillJ'; I,' ;\'11<1 whcre a completely continuous transition of armature current ,,11.1 I.l>r'llw is specified. This is achieved with another dual converter cir 1'1111., ::I",wlI ill Fig. 9.5 a, where both two-quadrant converters are conducting ::illllilI.:LII('()llsly. This calls for separate transformer secondaries as well as de mllplil'1': reactors because of the instantaneous volt age differences existing I)('Lw(:('1I Lhe two converters C 1 , C 2 • As seen from Fig. 9.5 a the armature CIII'I'(:IIL call assume either sign, '/." ,- 1. 1
--
(9.2)
l2
dq wlId iIlg on which of the currents prevails. The srna:llcr of the currents rep
a cirenlatillg current which flows throur,h I.he cOllvcrtcrs but bypasses 1.111' lo;\.( I brallch,
n :;;('lIl.s
IIliIlUIJ~)
ii :.flolll.! [)(' kq>1. a:: I " ' ! I('/,I V"
I'H\V I "
(9.3)
1.,.
[' "IIII':!
1111,,111 a~; ()II
145
9.1 DC Drive with Line-commutated Converter
1'":i::iI,j',, 1." 1','<111 ('( ' !.II(' ;I:::;o('i:~t.(:d n.d.ivl: alld 1.1,,' ,>1,.111'1' 1i :11I,1, 1.1" , ,·I,' ·u,l n l.illl'. ('1111'1'1'111. ::hould h('
L(7RC
ii
ú>,
ú>1
Lo ia
\.:J
-
e
L(7RC
e
C I Rectifier
C I Rectifier
C2 1nverter
C2 1nverter
C I Inverter C2 Rectifier
C I Inverter C2 Rectifier
ia" mM
i2 -------'
a
b
Fig. 9.5. Reversible converter with circulating current supplying DC machine
continuous so that there are no unnecessary waiting intervals delaying possi ble control action, The operating states of the two converters are indicated in the four quad rants of the torque/speed- plane in Fig. 9.5 b; for mM > O, the armature and circulating currents flow through C 1 whereas the auxiliary converter C 2 carries only the circulating current; the inverse is true for mM < O. To keep Lhe circulating current at a sufficiently low leveI, such as 10% of nominal cur rent, the mean voltage of the auxiliary converter must c10sely track the mean voltage of the main converter which is determined by the armature of the Illotor. If this condition is maintained all the time, both converters are active ;wd ready to accept the motor current in a continuous transient, without any waiting interval. This two-variable control is effected with the help of the two lirillg circuits. The auxiliary converter must be controlled to satisfy the condition UDl
+ UD2
= Rc(i1
+ i 2)
~ O,
(9.4)
I . (~ . the sum of the mean volt ages should be approximately zero; however, Lili,; is not true for the instantaneous voltages UD1(T), UD2(T), because one III' Lhe converters operates as rectifier, the other as inverter. The two volt ages 1\.1111 th( ~ ir sum are plotted in Fig. 9.6 for three different conditions, neglecting ,·I)III1I1I1t.af:ion and internal voltage drops. Clearly at 01 = 02 = ~ the sum 01' !.lu' iJlsLanl.all<XHlS voltagcs assumes very large values eventhough the sum "I' I.he 1IH':t1l va.lues is :r.( ~ ro. To prevent I.hcsc large alternating voltages from I'II.IIJi illl ( I' )( ', 'HH iv,' (' II ITI'II L", Lhe c.Íl'clllal.il\l!; I:IIIT('IIL rcad.ors Dc' seen in Fig. 9.5 1 111(' II 'qu i n'.! , Si ""I> 1.111'1'1' i i; nlw ll,YH "III' whirh (,;I.rrit's 1.111'. slllall ('ollt.iIIIlOIl S
1,1(;
!lo Control of Converter-supplied DC Drives
II/II
...
a1 !: a2 !:
= 45 (X2 = 135 ~
;1 II I \ \ J
90 0
0
(XI
II II I I
9.1 DC Drive with Line-commutated Converter
I' I \ \.J
,i2Ref
,,
h I' I' j' 1 \ 1 \ 1 \ 1\ \ , ,I , 1 , 1 1 'I 1 1 ,1 ,1
: 'I '/ '~ 1
a2 =45
0
a1
,
~
,,
0
= 135
0
rot!:
147
T
,,
,,
,,
..
~
iaRef
v
I, 1 \ 1
b
ii
~
UDt W Ref
$~
--1M Lo
W
rot = T
I"ig. 9.6. Voltages in reversible converter with circulating current
-LC
~I
~
U02
Lc
a Fig. 9.7. Control of reversible converter with controIled circulating current
cir<:lIlating current, these reactors need not be entirely linear; in fact, by allowillg the oue carrying the armature current to saturate, their physical size i:; redllccd . For smoothing the armature current ia, a separate filter reactor /'/1 IlIay h(~ induded in the armature circuito Io'or oht.ailling the desired current distribution for the respective main .11101 ;LIIXili;l.I·y-collverters, given a certain load current ia, one might think li, :; 1. 01 ali OpCll loop control scheme, where both firing angles are controlled 1"11I1.I 'y ;I,cmrdillg to 0:2 ~ 7r-O:l, which approximately fulfills condition (9.4). IIl1wI 'vC"I" , UI!: result would be far from satisfactory since the characteristics rv( .'l ) (Ir I.h(' Lwo firing circuits are likely to be somewhat different; also the CIIITI·1I1. d( ~ p c l1dent voltage drops would have to be taken into account. Thus I.Iw circlllatillg current would either be too large causing unnecessary los ses iII 1,11<' I.mllsfor mer, converters and reactors, or it would become discontin IIOII S, pr<: venting fast response, should the need for a reversal of armature CIIIT('UI. arise. The c10sed loop control scheme shown in Fig. 9.7 a promises a. vcry silllplc a nd effective solution to this problem; it is proven in numerous I\.Jlplications over many years [K24]. 'I'h( ~ idea. behind this scheme is as follows, Fig. 9.7: Based on the current 1('1(~I(,IIc(~ 'ia Ref prescribed by the speed controller, separate current references ,i I 110'1, 'i~ Ilo!" a.re derived for each of the converters. This is done by an electronic I'lIlId,ioll [l;ml C.I'ator, the characteristics of which are shown in Fig. 9.7 b; they lIT :oi O dlOse ll I.haL t.lre' eOllditioIlS 'I, 1
u"r
·1. :;' 1I, \f
'I,,,
IIl i rl( i l 11 ,,1, 'I 'J 11 0' 1 )
JI " f ,
i"
II ,· !'
(' 0"111 .•
are fulfilled, with the circulating current reference assuring continuous current flow. Thus the task of controlling armature current and circulating current is assigned to two separate current controlloops as described in Sect . 8.5. The circuit shown in Fig. 9.7 a can also be extended to include field weakening beyond base speed. With this control method the drive can operate in all quadrants of the torquejspeed plane allowing smooth transitions between the quadrants. How ever, this is achieved with considerable additional cost and complexity (trans former, reactors). Some savings may be possible, for example with continuous rolling mills, when the converter for reverse torque is only used for occasional braking and can be designed for a lower current rating; mOre substantial economies are made by turning to the reversible converter without circulat ing current that Was discussed before. This circuit permits also the use of the full rectifier voltage (o: = O), whereas with the circuit in Fig. 9.5, the rectifier voltage must be limited to the maximum value that can be supported by the inverter operating at maximum firing angle O:max. With today's state of con trol the loss in dynamic performance caused by the discontinuity at current zero is very slight, rendering the circuit in Fig. 9.4 the best solution for most applications.
1·lfI.
H.:l
!).
De
Contrai ofConverter-supplied DC Drives
D rives with Force-commutated Converters
("Ol! verter circuits dealt with so far have in common that they are fed Imlll siJlgle or three-phase alternating voltages; this is a prerequisite for phase c,,"l.rol cmploying natural commutation. As soon as a valve, having forward I)ias voltage, is fired, one of the supply voltages assumes the task of com 1IIIII.ating the load current from the previously conducting to the newly fired
'1'11('
vd.l v,:.
III a Ilumber of potential applications AC supply volt ages are not avail :tI,I", so t.hat this simple method of commutation is not feasible. Because a ,', IIld IIcl.illg thyristor cannot be turned off by electronic control, special pro visi" II s i Il the power circui tare then required to ensure that the thyristor to 1)(' li red has, at least temporarily, forward bias voltage and that the outgoing Lllyrisl.or becomes nonconducting and is briefly exposed to reverse voltage as :1. COlldition for blocking. A multitude of circuits exists which perform this "I'orcer! commutation". ln rccent years thyristors have been developed that can also be switched 011' hy suitable control signals while conducting (Gate-Turn-Off thyristors, ( :TO); tltis is a major improvement, even though the current gain for turn 011' is very low, requiring a short high pulse of inverse control current. These ,kvic,:s ltave in the last 10 years been developed to the point, where they (":1.11 n:place normal thyristors with several kV and kA rating in high power d.""lical.iollS, such as traction drives. 'I'1i,' 1I10S!. common switching device at lower power permitting electronic 1.11111 011' is (.Iw bipolar transistor. Since it can only conduct, when it is ac 1.1 \1.'1 1., 'd I,y ;1. I)ase current, it may be switched on and off at will, as long as i.l1(' :: W iI.eI I i I'1~ lasses are not too high and the transistor is protected against IIVI'I v"ll.ag':s caused, for example, by an inductive load. Bipolar transistors iII I':I.::CII( Icd Imm (Darlington circuit) are now available for volt ages exceeding III"" V ;llId Clll'rents of several hundred A, so that converters for medium size drlv,':-': II!, Lo about 500 kW can been built, but the design ofthe base control (·II'('lIil.s n:ll1a.ins cumbersome. i\ Ilol.her arca where promising developments are under way are field ef 1",'1. I.ransist.ors (MOSFET) for low power applications. These transistors are "llaracl.('l'isrd by very short switching time and, consequently, high switch· il'l', [J('.qllPIlCY beyond the audible range; being inherently volt age controlled, Ul('y ['('<["in: practically no steady state gate current, similar to vacuum tubes. Ilowcv(:r, caused by the relatively large gate capacil.y tlwrc is a substantial dlil.l'f~i IIJ'; C1ll'rellt which must be supplied by the conl.rol device for switching; :rI:;", ti", Oll~ st.a.te resislance is larp;er than wil.h hipolar l.ransistors. Mm(: n:c('lll.ly Insllbt"r! Gate llipolar "'n UIHi: iI.on-1 (1(; 11'1') lmve bepll de V,'loi'"d ii:; Ilil';11 (l0W('!' dcvic('s, COlllhillill/', 1.\[(' li i'l',h ('1111'('111. carryill/~ capaciLy "II,I!'"I;I.r I.ril.lI:: i::!.or:; wiLh Lhe :;illlpl,' ("""!.r,,l "r li,'l" dl','cl. I.rall:;i.~l.o n, iI.S wdl n: : Idr.h ::w ll.lllill /l, ~" )(>",I. 1<: 11'1":: l U '" ,,1'I'lll '"I.I ,\' 1.111' III,!:; L ,,1'()lIli:~i!l/ ~ :.; wi t.c1li 11/'; d" v II" '1< r, ,, Idl,.h,,,' Ili 'Wi>)' I1,p"lil" l.l.í" ll l1 . l'I'Ji ,'ld ll /( ilrl.1) 1.111' h V, "i\ I' I ./ ~ i,," . i\
9.2 DC Drives with Force-commutated Converters
149
further possibility, but still at an early stage of development, are MOSFET controlled thyristors (MCT) but their potential cannot yet be assessed. Clearly, the work on new and improved high power electronic switching devices is continuing at a rapid pace and a state of consolidation is not in sight. This strongly affects the development of the circuitry and control meth ods for converter equipment. ln this section it is shown with the example of a sim pie force-commutated chopper circuits for DC drives, how the new devices may result in better dynamic performance, simplifications and, possibly, cost reductions.
L
L
~o~--~~"~--,
ia
ia
Uo
Ua
a
Load
Load
Uo
b
Fig. 9.8. Principie of single phase chopper circuits
(a) Step-down chopper (b) Step-up chopper
ln Fig. 9.8 a,b two basic circuits for DC/DC choppers are seen for gener ating from a constant direct bus volt age with low losses a controllable direct volt age for supplying a DC load. The electronic switches in conjunction with an inductive energy storage are operated with a controllable On/Off duty cycle to produce a desired mean voltage U a across the output terminais; the circuits could be used as single quadrant supply for the armature of a DC mo tor. The electronic switch is carrying only forward current, reverse blocking is not necessary with the diodes included. As an example that is mainly of historical interest, a typical converter cir cuit of the type of Fig. 9.8 a, employing forced commutation with an ordinary thyristor is depicted in Fig. 9.9. Its purpose is to convert the constant direct volt age Uo to a lower adjustable volt age U a < Uo for controlling the speed of a DC series wound motor. This problem arises with vehicles, supplied through a DC rail or catenary 01' from a storage battery. As the thyristor TI which corresponds to the switch in Fig. 9.8 a cannot be turned off by electronic con trol, a resonant circuit with the auxiliary thyristor T 2 is provided with the purpose of creating a zero intersection of the current iTl, thus facilitating the blocking of thyristor TI' When using a GTO which can also be switched off electronically, the same effect could be achieved more directly. T lw 1)(:/1)(: (,()Ilv(~rt.er operates a.s a chopper, which, by means of an i,ll'drillllC ::wil.rll, «III1\(~cl.S th(~ illdllctiv(~ load with the pa.rallel shunting diode /II 1", r! "dicillI.1' LI) 1.11(' cOII::/.:!.IIL ";lIpply voll.:"';i>. III' varyiug Lh(~ dllty cyrk, til('
l i .!)
'L
9.2 DC Drives wíth Force-commutated Converters
Control of Converter-supplied DC Drives
kK----- _-
Tum-off Thyristor (GTO)
a
......,.;tVI
~
1- - - - - -T- - -. - - ,
Ro
. fJ
'Tf
Re
or
Uo
I I I I i Df L2 I ___ .J
--1--__
-"'" 1__ O2
ua
l;e Ue ~--zí:
J
Fig. 9 .9. DC/DC chopper supplying a series-wound motor
J
~t2
\----,
"
/'
ID f
/Cctificr
.
1,\ iTl
-,
I
I J,lttery
- ____ ....
........
I I I
Lo
ia
--- -
Thyristor with force tum-oft I
151
t
~t3
I
t tI F;ring 01 T,
\
\Y
V
j t,
r--ZL= /~;
\I
voltage U a is changed. The switching frequency is chosen sufficiently lIigh (> 100 Hz) so that a relatively smooth continuous current ia flows in the load circuit. The converter shown in Fig. 9.9 is of the single quadrant variety, wilic\) is adequate for a series motor but with additional components it could 1)(' ('X I.(~lldcd for two- or four-quadrant operation as well. The circuit should ollly \)(' collsidered as an example; there are many different versions having ::illlilar proper Lies. The operation of the DC/DC converter will be explained iii :: I.,'adY Klatc where ali variables are periodic functions with period T . The (' III v(':: iII I<' i!(. 9.10 have been obtained by a detailed simulation of the circuit 1 111 :1. d i/';iLd (,o ll1puter. '1'1 1<' IIlotl e of operation is briefly as follows : The main thyristor TI is I)('Ji odi c; dl y Jircd at time tI + vT, v = 0, 1, . . . connecting the supply volt age (I" 1.0 1.11, · i11dll ctive load circuit, ua(t) = Uo, at the sarne time blocking the :illlllllill! '; di( )(le DI which had carried the continuous load current i a. This (IIII.' ;,·S i.llC pos itively charged capacitor, ue(td > 0, to be discharged through 1.11<' J'( ~S()Ilitl1t. commutation circuit consisting of Le, L2' R2' D 2; after the first Ilall' wav(~ t.his transient is interrupted by D 2 as the current is about to change il.s ;;ig11 , II'
( .~ "'/1" . di
1 "1 •
( ~ Io ~i )
iT2
----""\
t t2 - T
----"'\ \
2
1 Hring 01 T
íV02 . . , !., ,~,I01, ----j v',
,_,___ -; r·,
Fig. 9,10. Símulation results for DC/DC converter circuit
The bias volt age across thyristor TI eventually becomes positive at time t3, when U e passes through zero; in the interval t3 - t2 the recovery of TI must have been completed. The rechargingofthe capacitor continues until Ue = Uo, when the freewheeling diode DI takes over the load current and T 2 reverts 1.0 blocking condition. The electronic switch then remains in this state, with Iloth thyristors blocked and Ue > O, until the next switching cycle begins. ln Fig. 9.10 the commutation interval has been enlarged for better visibility. Clearly it is criticaI for the functioning of the circuit that the alternating volt age of the commutating capacitor is of sufficient magnitude to ensure an adcquate hold-off interval t 3 - t2 when the main thyristor is reversely biased. I[ this time becomes too short because the charge on the capacitor was in ~i il{fic ient or there is excessive load current, TI will refire and a short circuit colldition develops which can only be cleared by a fuse link or a breaker. ln I,"is r< :sp ect a commutation failure of this circuit is more serious than with a l i ll<'~ <:O lllJ\lill.al(:d converter where the situ a tion could be resolved by advanc illf', LIli' 111',1, firillf'; illSL;Ult.. 'T'hi H d al1{';(~r 01' l()s ill ~ cOl1t rol over the converter
I r,2
9. ControI of Converter-supplied DC Drives
9.2 DC Drives with Force-commutated Converters
('xists with alI thyristor circuits employing forced commutationj it can only Iw avoided through careful design of the circuit and by having a fast cur l('[ll. control which prevents excessive load current before it approaches the ,'01 II II Illtation limito Âllother detail is seen in Fig, 9.10 at time t4 where the capacitor, having 1)(" '11 clIarged beyond Uo due to the effect of the source inductance Lo, returns ;;"",,' 01" its charge to the battery through diode D 2 j since it is desirable to h:IV(' I.Iw capacitor well charged for added commutation capability, this loss ,,j rh:I.II~(' SllOUld be prevented by using instead of D 2 another thyristor that i:: I i1'1'<1 I.o!-\cther with TI. '1'11(: ,~witching sequence is repeated at a frequency f between 50 Hz and IWlllaps 1 kHz, depending on the type and power rating of the thyristors. '1'1.(' control effect is achieved by pulse-width modulationj for example, if Lhyrist.or TI is fired at the frequency f with a fixed phase, the On-Off ratio is all.ned by advancing or retarding the firing instant t2 of thyristor T 2 . When 1I"I;I( ~clillg the commutating transients the volt age U a can be approximated I',l' a periodical square wave having constant amplitude Uo and variable pulse wi
V'a
t2 - tI
< -- = - T - < 0.9 Uo
ing branchesj nevertheless, digital simulation is an indispensible tool at the design stage. Naturally, after the converter design is completed, much simpler models can be used for designing the control system. An extensive specialised literature exists on De/De converters describing numerous details, particu larly with regard to commutation circuits [3,48,51).
lL iSf
W
ia
Load curve WRef
(9.6)
1.11 i," avoids the large power losses that would occur with a series resistor. UI" rOIlJ"se , ali components in the converter circuit involve some losses which hil v,' 1.0 h(~ \; Xalllilled very carefully at the design stagej still, the over-all dliri"I\(,l' uf ;t I.ypical De/De converter may be better than 95%. The limits (I ~ II) )" alie! (t2 - tI) -t T cannot be reached by continuous pulse-width (", 1111.1'" hl'c;I.IISI' some time is needed for the commutating transients. 1"'(1111 I.Ili,~ silllplified description it is apparent that the analysis and, con 1:'·'1'"'lIl.Iy, I.he d(;sign of a converter with forced commutation is much more , (1111 pi", I.h;\'1I t.hat of a line-commutated converter. The volt ages and currents, 1,,1 n:IIIIJllc I.hose directly affecting the thyristors, can only be determined by 1.('!lII)IIS Ilalle! calculations in the successive intervals, by measurement on a l:dlOlill.ory IIl1it or by simulating the complete circuit on a digital computer. VI'I,\' dJici(:II1. programs have been developed for this purpose which require "1I1 ,l' :1. t.opologi cal description of the circuitj from this information the differ I'lIl.ial ('qllatiollS are automatically generated. With given initial conditions, t.Il,' 1I11111nical illtegration proceeds until either a control input occurs (e.g. lil'illl'. 01" a Ll,yristor) or one ofthe components reaches a discontinuity, such as iI. di,,,[P hlockiug, at wlIich time new equations are cldillcd (or new parameters "III.I'I'C
L
GTO
j
Uo
Mmax
I
I
a
01
Ua
De chopper
I
I
j
153
CD
b mM
Fig. 9.11. (a) Simplified drive control scheme of drive with DC/DC converter, (b) steady-state characteristics
The De/De converter, like the line-commutated converter, is a power amplifier with fast response having limited overload capacityj hence it calls for dfective protection which is best provided by closed loop control. With vehicle e!rives which are often coupled to a large load inertia this is of particular illlportance because acceleration should often take place at a chosen current limito Therefore cascade control having an inner current loop is well suited ;l1so to De/De converters. As was shown in Sect. 8.5, very simple dynamic modeIs are quite adequate ror the design of the control system, provided the innermost loop contains I,csicles the switching converter - a simple control plant with low-pass char :t,c.(",(,ristics. When neglecting the commutation transients and assuming con I.ÚIIIOIlS load current ia, the converter in Fig. 9.9 looks from the load side like ; \.11 illlpressed voltage source, where u a alternates between Uo and zero with variablc duty cycle. Hence the circuit can be represented by a mechanical :, wil.ch as seen in Fig. 9.11 a that may be operated by an On-Off-controller It:t.villg a lIarrow hysteresis loop of width 2L1. With the real converter, the ,', 1I11.rolkr woulcl issue firing pulses to the main- and auxiliary thyristors. TIl(' hlock diaf';ram of a series wouncl motor, Fig. 6.3, is quite nonlinear, 11111. wil.h ali ()II ·(){J"-coltt.wll(,r for (,11\, illlwr CIIJT(' lIt control loop no prob 1"111 1: ''',' "111'011111."(1"1. '1'11(' :;yHI.\'1I1 1',',11111.:1 iII a, "'"i1,;i ·Iillear illIWJ" dos('.d loop
11, ,1
~) .
( :() IILro] of Converter-supplied DC Drives
9.2 DC Drives with Force-commutated Converters
"0 I lu
' ', I~~~~~~~~~~~~~~~~~~~~ ~ ~ ~ ~ ~ ~ ., ,: , ~ l,r)
t
1 m,
'''~
: m)
.,
"l~~ I" i l-t, 11,12,
Silllul nted starting transient of drive with De/DC converter
wil.lt :r.ppro xilllal.c!y unity gain. ln the superimposed speed controlloop the 1I11'{ 'II IUl ic:d illLegra tion is added (neglecting friction), however with variable I '.I U II, 1 H' (, ; LlI S (~ Lhe torque of a series motor is determined by Peia :::::! i ~. The " '111 111, W()lIld Iw different speed transients depending on the operating point 01 !.III' IlIOI.UI'. '1'0 avoid this effect, it may be advantageous to add a nonlinear 1'r 1l H'l.i oll g( ~ l wrat or producing a torque signal, '/II' M '/}/.)
IUleI
f
(~) Zal
,
ia
~
O,
(9.7)
an innel' torque controlloop instead of a current loop; no partic Irlal' n.C(,lIJ'étcy of the function generator is requil'ed. The result is seen in the Irlor!, di agl'étlI1 Fi g, 9.11 a, where all nonlinearities are now contained in the illllc r loo p. Tl w pulse-width modulated waveform is l~c n cratcd by the closed 1. ( IIIJlI<' COIl \. rol loop wit!tout tlw need for a li o lC \.,~nlld dock; the frequency ,'lt[l III' itdjll sl.(~d uy t he dlOicc of t,he hys t;(~reHi:4 wid l. h. J\, ~aill, wi t h a simple ()II/( )Jr.I' o H l. rlllkl' , \.he I' wiLchi ltg' fl'< ~qlJ(~ II CY III' UI<' ('oll vn\'( :r is not cO)H'i t a nt I lI ll. . ' xlt ih it.l1 ~~ II IIU/ i llllllll iII t.lu ' ("'n t, l'I \ III' 1.11.. ' " (ll d,I'ol n LUI.(I' rlRI, huI. t.his , ' III IIII I' H 11,0 )1lt1'l 1l / Ul I " I ~H l U! Lh l' I 1.1. i II Í'lu 11 11,1 rl " ''1III '' " I ',~t 111 Il ldl i ,· i.l' IIUy hil':h, Ilow " V( 'I, wl" '1I /•• ·.. 11111,'11 11 /Iw lt.d IÍ 1'I '; r l'l ''1 I1 ' ' I I(' V 1/1 11 1" ','111.-.1 1." fLV" irl 1111."1 r"I ' C' IIC ' C' Lo rOl'lll
155
with signalling circuits laid along the railway tracks, a constant frequency pulse-width modulator is preferable. If the vehicle contains several indepen dently fed drive motors, the firing of the converters may be synchronised out of phase to increase the ripple frequency of the current in the supply line. Figure 9.12 shows a simulated starting transient with speed-dependent load torque; the operating point in Fig.9.11 b moves from 1 via 2 to 3. The inertia of the drive was chosen very small to clarify the details of the switching operation; with an actual traction drive, the acceleration phase may last many seconds while the switching takes place at several hundred Hz. When selecting the parameters of the controllers, sim pIe design methods are again applicable, for example as discussed in [38J. As mentioned in Sect. 8.5, On/ Off control can be treated more accurately as a nonlinear discrete control system which however would involve considerable complexity; in view of the good results with the approximate methods this is normally not justified. As indicated in Chapo 6 the use of series connected motors is restricted to only a few applications; for high dynamic performance drives, such as reversible servo drives, De motors with separate excitation or permanent magnets are normally preferred; they require converters with four quadrant capability in the ia/ua-plane. The bridge type converter shown in Fig. 9.13 a is connected to the ar mature circuit of a DC motor; it may be supplied with constant volt age UD from a DC bus or a battery. The converter contains four electronic switches where two in each half-bridge are drawn in the form of a transfer switch (at the sarne time excluding accidental short circuits of the DC bus) ; the diodes which can be part of the electronic switches allow an inductive load current to continue during the short protective intervals, when alI contacts are open (similar to the red-light-overlap on a signal crossing). By assigning logic symbols SI, S2 to the otherwise ideally assumed switches, the voltage equation of the load circuit is Za La -
dt
+ R a Za' + e =
(9.8)
Ua ,
where, depending on the switching state Ua
holds,
1 = 2(S1 - S2)
UD
and
, ZD
1
= 2(S1 -
,
S2) Za
(9.9)
I r,j;
9. Control oE Converter-supplied DC Drives
9.2 DC Drives with Force-commutated Converters
iO
157
SUpp/y
2
c
La
3 io
Ra
4-Quadrant converter
$2 -1
U
o
ej
ii
U
a
C
UD __
I)
~/_
-I
I/ua
, S = -1' II..., ......L .......L._
T
S",.S
I~: ~o_
o r-:-J-_ U - - - U- a
=r:;....::::... / I
T
S/i;
Ir
1...--.....1_
Symmetrical modulation
- U 12. _____ 2
~
__ _
Cyclical modulation
inductive load,
TIl(' pIIIs('-width-rnodulation (PWM) of the converter at the frequency I j'1' (';tII fullow different switching strategies, as illustrated in Fig. 9.13 I, w; I.ii 1.1I(' oIII. P III. vol tage U a during a switching period:
f
• \ l"ipol;11' IlIodlda.tioIl (1111'
,':y III 11I1'l.l'i(' ; t! Illudul ation
W iI.l, ,," i.~ .l lwdulation pattern the switches are operated in diagonal pairs, S, :-,"2, so tltat the short circuit interval is ornitted and the output vo ll.;1.1 ';<. ;dl.el'llules between the values UD and -UD. During the unavoidable (I" I I. iII I'·ig. 9.13 neglected) protective intervals the diodes are carrying the 1";1.'/ ( ' lIl'reIlL
Illodulation I.lri s lIlodulat.iOIl scheme the two transfer switches are operated se IJI[(·III.;all 'y, so I.hat. th e output is alternatively short circuited at the upper 01' lowcr sllpply bus. H(~ llc(~ the output voltage U a assumes a ternary wavc 1'01 ' 111, '11." '/I.u, O, " · ltf).Wh(~ n~as witlt symnwtri ca.l switchillg oldy Llte mean II I' Li", olll.plll. voll.a!,,(· ('.<1.11 I)( ~ cOllLrul!cd iu st.c;l
• (:y d ie
l'I i lllll ll ll .l l1l'.
J
~ir-J···~,
FiJ.C. 11.13 . Four-quadrant DC/DC converter with (,, ) (:il'('uiL, (b ) Different modulation patterns
1;,1'
Ta
- - -"--r
~1
-~
Ulllpolar modulation
Uo
Iiii 1'1 11011 i( ' H III' 1.11(' Olll.plll. volta",..,
'11." ,
Fig. 9.14. DC Servo drive with voltage source DC/DC converter
A four-quadrant converter with lGBT-switches is depicted in Fig. 9.14 as frequently used for DC servo drivesj protective circuitry is again omitted. For simplicit.yan OnjOff device is drawn for current controI but a linear current controller with constant frequency PWM would normally be preferred. A diode rectifier followed by a smoot.hing filter, whose capacitor C ab sorbs also the modulation-induced ripple components of the link current iD serves as the supply of the DC link with constant voltage UD. For instance, when rapidly braking the drive, power released from the kinet.ic energy flows back into the DC-link causing negative current iD and, because of the uni directionaIline-side rectifiers, couId result in an overcharge of the capacitor; this is prevented by dissipat.ing the energy in a resistive ballast circuit that can also be pulse-width-modulated, depending on the link voltage. ln view of the losses this is only practical with srnall drives or when it happens only occasionally; otherwise a reversible line-side supply (an active front-end con verter) is preferable as will be shown in Fig. 9.18 and further discussed in Sect. 13.2. Some of the steady state waveforms in a converter like the one in Fig. !J.14 are indicated in Fig. 9.15, showing the output voltage U a alternating hdween UD and -UD and the alternating current components of iD, which lllllSt be absorbed by the capacitor. The control can be arranged as beforej il.U illller cnrrent loop controlling the converter via a pulse-width modulator is illlportant. for safe operation. The current controller in Fig. 9.14 is again drawl1 ;IS
I!, fi
'l. Control
r.
of Converter-supplied DC Drives 2f,.
0.1')
'.0)
/l1' ~/iYlfPH21f3 I~Ref
""II
9.2 DC Drives with Force-commutated Converters
159
Phase shift (Ual. ial )
Ua
n
Jc1 c1 c1 c1 c1 [L o o o o O":
1r1 c1 r1 r1 r1" ;
i
U U U U U
U;
I~ il; . tI.15. Waveforms of DC/DC converter with symmetrical modulation .
al'pJicaliollS this justifies the assumption that the current controlloop acts as colll.rollable current source having instantaneous response. Current limit i" ac:bil ~ vI~d by limiting the current reference produced by the superimposed :iJ >('('d ('()lIl.roller. The next higher levei of control could be a position control !\lO p :IS showlI in Fig. 15.9, where the response may be further improved by 1" "<1 J'o)ward siguais from a reference generator. '1 'r;I.II ::i:; l.m e<mverters have the important advantage that they can be li wil.d", 5 kHz, thus enlarging the control bandwidth as '''"11 " 11 ,,<1 1.0 lillc-commutated converters. With field effect transistors or I( : I\'1' 'i:, I.hl ' rl(~quency can even be increased beyond the audible threshold 1(; !t 11 '1" so t,hat the drive is no longer emitting objectionable acoustic noise. ' ! 'II(' II>II/' quadrant converter shown in Fig. 9.14 is also suitable for produc ill/', 1\( : iII Lhe form of pulse-with-modulated voltages and currents, provided 11.11 J'n ~ qll('lI('.y is sufficiently below the switching frequency of the converter; i.l ti fi i.~ SCI~II iII Fig. 9.16 with the example of approximately sinusoidal out P"l. 1'111'1'1 ' 111. 'i " in a passive (e = O) inductive load circuito With the assumed (),, / O!!' colllroller the modulation is symmetrical and the switching frequency \ll lIil~S !ln!. Lhis could be remedied with a PWM mod1l1ator and fixed clock r/" ' qlll ~ lll'Y, Ali illspection of the curves reveals that the <:onverter passes cycli ",,!l y I, h011 i', h
rlnnnnnm~ij~~~~nn ~;1r1(]nnnon~~~~~r~~~u~4\r1~ ~ ~ i ~ ~ m~~[~~nUD' -UU~ij~~~~~ io
n
• I
Fig. 9.16. Waveforms of symmetrically modulated four-quadrant converter with AC output
• Cylindrical motors with a slim rotor hence low inertia, but otherwise con ventional design and • Disk motors with axial magnetic field and radial armature winding, which is directly printed on a fiber reinforced disk; the brushes are sliding on the printed conductors of the commutator. These motors are charakterised by a very small armature time constant and low inertia; typical data are given in Fig. 14.1. Because of the progress with controlled AC drives, servo motors are today mostly synchronous motors with permanent magnet excitation, Sect. 14.1. When discussing phase controlled converters in Chapo 8, the problem of line-side interactions by harmonics and reactive current was mentioned. This applies also to uncontrolled rectifiers, such as in Fig. 9.14, particularly when a simple capacitive filter without smoothing choke is employed which causes line currents in the form of short spikes at the volt age peaks. With the growing use of power electronic equipment these effects are now receiving increased attention in technical standards and regulations. With force controlled converters the reactive line currents can be greatly reduced while the harmonics are shifted to a higher frequency range, where it is easier to suppress them by filters; at the sarne time the problem of reverse power flow can be solved. This is shown with the schematíc diagram in Fig. 9.17 of a single phase converter which is of interest, apart from low power drives, on AC traction drives with intermediate DC link converter (an extension to three phases is di scussed ill Sect. 13.2) . The circuit is essentially an inverted four-quadrant I~OIlV(~ rt;t: r iII ),'ig , 0.13, where the inductive impedance LN is now represented II,Y i. 1!<: ka!t:\I';t: 01' t.1!I~ Iilu:-sirle tra nsformer. The simplified equations are
11111
(:olltrol of Converter-supplied DC Drives
!) ,
9.2 DC Drives with Force-cornrnutated Converters
ir 'Nb
y
controller. When shifting the current reference by to with respect to the line voltage UN, a reactive component of the line current results. Caused by the single phase supply, periodic fiuctuations of the link voltage UD at twice line frequency cannot be avoided but are greatly diminished by inserting a resonant círcuit tuned to double line frequency, as is indicated in Fig. 9.18; this is the usual practice on traction drives. The load-side DC link current iDL acts as a disturbance to the voltage control, its effect may be reduced by feed-forward to the current reference.
!lN
'NV/
',':!
UN(t) i
ON
k
lN
~;
i it)
1'15
iit) = bN(t)
2
-1
- -- -....
-.-.
5hort circuit
Transfer
interval
Interval
51 =52 =-1
5 1 =1,52 =-1 A
UN(t + to)! U N
Jo' i ~ . 9.1 7. Principie of a single phase line-fed four-quadrant converter for supplying ;) 1)( :
lillk
iN
'/'N
L N -
dt
+ RN
,
ZN
=
UN -
UB ,
(9.10)
iNRef
iOL
;1,lId
,Á'I/,f) dI.
, ' = ~DN
,
Uo
(9.11)
-ZDL,
F ig. 9.18. Control of a single phase line-fed four-quadrant converter
wlwJ'\ >'i/>/, is the load-side link current; furthermore 11./1
.: (,<" 1 -.52 )
uD
und
iDN
=
~(51 -
u/ V
82 )
161
iN
U
(9.12) 400
f---
UN
I" del ::, II pprw::illla tely sinusoidalline current i N can only be achieved if the
1)(: lildí, 11:'t l1I lhe line voltage is much lower than UD, i.e. near the zero in I.I 'I'fil >('l.ioIlS oe 'U'N; hence it is necessary, as shown in Fig. 9.17 b, to create ', t'rll i Ill.crv;ds of the voltage UB for storing in the line-side inductance energy I.Il il.1. i:, slli>sequ(~ ntly transferred in the form of a current pulse to the link I'II,p;u 'iI.OI, "'hus it is necessary to employ a unipolar or cyclical modulation, w I w J'\ ~ I.hc COllverter bridge is temporarily short circuitcd at a DC bus. '1 '11(> switchillg frcquency of the converter is best chosen as an integer 1IIIdl.ipl(~ II of the line frequency to avoid in st ead y st.at(~ sllhharmonics of the l i lll ~ 1'11/'1'<:111.; with a 50 Il7. :mpply and a clode fn :IlIIl'Jlcy ror ius ta.nce of 2 kH z, n til) swiLchillf,'; (:yd(~f> wou ld fit illto one lillC voll.iI,l':(> Jlcriod, rcsultillg in a r.d rl.v :n llool.lt w;l.vd<mll ()f Lhe ('.111'1'<:111:, II pll:-ls i lll,' co//Crol ;'i dlCIIll! i;{ :, !t UWIl iII I" i,,;. n,I'H, whl'l'tj II. volt,ag., cOJl t rol 1""11 I~ II' '1/,11 iH rHlpI'I'illll)(IH,'d OH ali iIIU('l' " UITI' fll ''' ('III 1'\11' '/:N ,- '1'1-1\' I'IIlTI 'IJI. ,, 'I,'rl' II/ 'I' ; N 11 ,, 1 i!:1 d\,ri VI'd 1"1'0111 t,h" l i "I' vll lt HI'.~ ' ffll ~11, I,nilli l,,'; Il. Hil\ w ,,,i d a l Ii/\I' " 11111"111. 11'(" " ' 111 "' " !-III' IlI ftl\ l.! I.I\/'" ,II' wld, I, ln dnl"I'lll hll'd I, y LI ... v.. II';"F,"
o
11\
i OL
1\ A A A A 1\ A A
200
~lk\IViA\IViA~Vlk ~ i -200
-400
"';g. tI.lU. Si lllllhL"d loa
1\
lillc-fecl fOllr-lJ"adrant converter
Ifi ;~
!I .
Control of Converter-supplied DC Drives
1\ ~iiIlIlIlated load reversal is plotted in Fig. 9.19, initiated by inverting the
which results in a phase reversal of the line current in steady ::Iil.l c, i. c. feeding power back to the line [13]. The resonant circuit shown in leig. !l. I 8 is included in the simulation model, thus reducing the harmonics w ut.<:nt of the link voltage. I"IIITI'tll. i ,lL ,
10. Symmetrical Three-Phase AC Machines
The asynchronous or induction motor is the most widely used electrical drive motor; its invention at the end of the last century has given a strong impetus to the transition from DC to AC in the field of generation, transmission and distribution of electrical energy. Its main advantage is the elimination of alI sliding contacts, resulting in an exceedingly simpie and rugged construction. Induction machines are built in a variety of designs with ratings from a few watts to many megawatts. Unfortunately, the speed of line- fed induction motors cannot be continu ously varied without additional equipment or without incurring heavy power losses. Even though the problems of efficiently controlling the speed of induc tion motors have been investigated for decades, ali solutions realisable until some years ago were unsatisfactory with regard to complexity, efliciency, dy namic performance or cost. It is only due to the progress of semiconductor technology in the last 30 years t hat suitable static frequency converters can now be built at acceptable cost, making the induction machine a versatile adjustable speed drive for many applications. The substitution of the mechanical commutator, by a power electronic, solid state converter reduces the AC machine solely to the task of electrome chanical energy conversion, resulting in much higher power density per vol ume or mass. This is exemplified by two main line traction motors , the out lines of which are shown in Fig. 10.1, together with some characteristic data. The motor at left is a proven design, operated for many years in Intercity locomotives; it is a single phase series-wound commutator motor as men tioned in Chapo 6, voltage controlIed by transformer tap changer. At the right is a modem AC induction motor, fed and controlIed by a GTO volt age source converter. Clearly, the new motor is superior in every respect; it is also suitable for high dynamic performance torque control, thus alIowing ful! utilisation of the adhesive frictional forces between wheel and rail. As a result, a locomotive containing four of the new motors can produce a similar traction force as the former much heavier locomotive with six single phase motors, the torque of which is pulsating at double the supply frequency. Thc tbcory of the induction motor under dynamic conditions is somewhat iuvo!V<'d IlJ :citlls(: of t.he rotating rn af!,lId.ic lidrls , the spa tial relationships of whiclt dl'Jl<'lId ou :: 1" '1:<1 nlld loadj i!. .. 11.11 Iw In':Ü' :I[ h c l'(~ only in silllplifif~d
111 ,1
1II.
1', >I'" \, () II tlIe other hand, the equivalent circuits, derived for steady state "1)('r:ll.ioll with sinusoidal volt ages and currents, are inadequate when dealing wiUl I.rallsients or when the motor is supplied from a switching converter. ln I.h(' ['ollowing, the steady state condition will be treated as a special case of Uw dyuamic case. The mathematical model to be used is tailored to the needs of controlled d ri ves. It incorporates most of the qualitative features of an actual motor lilll. would not, of course, be accurate enough for the purpose of designing !.II(' Illaehine.
1.-:1 -
---,.p.
b)
S"ri e:~-wound
Lnll kW, 7730 Nm
Conto rating
,d, 1~):lIlJllin - l H~dll NIIl
IlilUI
III;"
:I ~ , ~ ,II h 1" 1~~ II 1'1 '. ' >I ;:
920mm
Three-phase AC induction motor
tator motor
1428 kW, 9155 Nm at 1490 win- 1
Torque (5 min) Max, speed Mass Inertia
11600 Nm 4200 min- 1
a
2660 kg
0 s (ex,t)
22 kg m 2
,
F I,( . 10 . 1. ( ;lIlllparison of traction motors (Siemens). (II ) S ' "I·.I,· I'I"\.~,, o;eries-wound AC commutator motor, ( I, ) 'l'1>I ' I"'-pll;c~e AC induction motor
a.
I
ex
-TI
b ~s(ex-4TI/3)
S -TI
zs:
~s(ex)
'S:6~SV
~s(ex-2TI/3)
zs:
3
TI7KZ
c
10. 1 Mathematical Model of a General AC Machine I L i:: ...sslIllwd that the stator S of the machine is a hollow iron cylinder with ,.; rl'lda, (TOSS sectioIl, containing a concentric rotor R $0 that a llarrow airgap "I' I·{)W;(.alll. radial length h exists between the SIlI()oth cyJillctrical surfaces to wlll<,h sYlIlIlIl'I.ricaJ (.!trpc- plta:;c winctings of ll('I ': lil~ihk JI(~ight, arp assumed to III' :d.J.;u:lwd; t.Iwir sJliLl.ial illllperetlll'llS c1ixtrillllli,,":: IIlay h(~ thOllght o[ as IlI'illl·. prodll('('d hy slIiLallly plal'cd I.l,ill sLI'1\!,,\:J "I' ax ia l ('Illldlldol's or COll ""II'Liv,' :d,,·..I,::. Illlth 1I( ·nl.mls "I' UII' :iI,i1,r· "()IIIl, ,,, \.,·d wi lldilll (:-l .U:(· isola!.(·", l.\w
" I' 1,1, (" fOI." .. willdÍlIj': :U',· ,·,iI.ll<'l· "",,111" ' 1.,,01 I.,! ::1 i l> l' i "I'':: 1)1' :;b,,1'1. ('ir t"Ilíl.,·d illl',"1 " II II ,v. N :ó • Nu 11.1 '1' 1.1... 11111111""' ;0 "I 1,,11 Iti !...!1 1.111' 1111 1111 '1 11''' """:«' (.I ·,I·I
165
winding.The permeability of the fully laminated stator and rotor iron is as sumed to be infinite; saturation, iron losses, end-windings and slot-effects are ignored. The following discussion applies to two-pole motors; with multi-pole machines the synchronous speed is reduced correspondingly. This effect is exploited with pole-changing windings, however at some loss in utilisation of the motor.
§f,omm
11200 mm
,I) , 8 70 mm
!:1I11I1l11I
10.1 Mathematical Model of a General AC Machine
Symmetrical Three-Phase AC Machines
.... '/;71 '
ti
- TI
.. a
.........."',\
o
...
,
Fil!;. 10.2. Symmetrical AC machine
CrIlSS ~cctioll wilh two-layer airgap windings,
(I.) {lJ)wrapp"d slat.or windings,
(t') 1\"'1)('1'1'1.1"'(\1-1 di:·ll.rihIlLioll or xLa.l.nr willdillgs,
(Il)
(01) 'I'rllVl'II ; "I',
II.IIIJlI'II'I.III'IIH W:\VI '
ntll~(,d h.v
t.Il1" ·'·' I''':''''''
I'.II ..
r<:"L~
I{jG
10. Symmetrical Three-Phase AC Machines
10.1 Mathematical Model oC a General AC Machine
The variables are defined in Fig. 10.2 a. a is the angular coordinate in the stator wíth reference to the axis of stator winding 1 which is indicated by its central turn. The centres of the identical windings 2 and 3 are positioned 0 at a = -y = 120 and a = 2 -y = 240 0 respectively. Corresponding definitions hold for the rotor windings, where f3 is the angular coordinate, again with rderence to the centre of winding 1. é(t) is the angle of rotation of the rotor, Il)(~asured in the stator frame of coordinates, w(t) = dél dt is the instantaneous ;tIlgular velocity of the rotor. The magnetic field in the airgap of the machine has radial direction be LtllSC of the smooth parallel stator and rotor surfaces and the postulated irdillite permeability of the iron; since end-effects are neglected, we are deal illg with a two-dimensional magnetic field problem. The three stator currents isl(t), is 2(t), iS3(t) may exhibit any waveforms; I'()r: r(~asons of symmetry all currents are retained throughout the calculations, CV(~1l though one of the currents is redundant because, due to the isolated Ileutral, 'is1(t)
=O
+ is2(t) + iS3(t)
where is(t) = isl(t)
+ is 2(t) eh + iss(t) ej2 "(
(10.5)
is a time-dependent current vector in the complex plane perpendicular to the motor axis and is(t)
= is1(t) + is 2(t) e-h + iss(t) e- j2 "(
(10.6)
is the corresponding conjugate complex vector. The ampereturns 8s(a, t) are of course real, being a measurable physical quantity depending on a and t.
e jy
(10.1)
is:ft) !\/isit)e
is valícl at any instant, Le. the currents are balanced; this applies also to Ll _ COllllccted motors.
+ is 2(t) '!9(a - -y) + iss(t) '!9(a -
2í'r 2 -y)] ,-y= 3'.
J2y
2
N 8s (ex,!)
'S2(t) e fY
Wit,h these simplifications and definitions, the radial ampereturns wave 01' t.11( ~ stator windings consists of three spatially sinusoidal, stationary terms, w li ich a]'(~ modulated by the currents (-)" (II'. I) ' N S [is1(t) '!9(a)
167
s
Mt)e a
-ja
b
Fig. 10.3. Complex current vector
(10.2) !\\mrdillg to Fig. 10.2, 8s(a, t) corresponds to the stator ampereturns en <"1",,: ('<1 I,'y ii. ra.dial magnetic field line crossing the motor at angle a; because "I 1.1)(' ;1,SSIIllICd high permeability of the iron these ampere turns appear as III;q ~ "d()ltIotive forces at the two airgap crossings. The nondimensional spatial I'rrlll'l,i()1I - 1 ::; '!9(a) ::; 1 in Fig. 10.2 c characterises the ampereturns distri 1'IIt.io/l ()f one phase of the stator winding, it could be written as a Fourier ::ni(!:,; . Wh cn lhe windings are fed with alternating currents, each term in Eq. ( 10.:J.) o~cillat.es as a standing wave, fixed in space; their superposition results iII ;t, travdling wave 8s(a, t), as seen in Fig. 10.2 d. The spatial harmonics rllH.y he dÍluillished by reduced pitch and skew of the windings, resulting in /1 ( II') ., cos cr. I ty illl,roducillg rornplex llotation, :I~ ( ( ,:in
('OS IY
IU ld
'l (.
:lU)
,
etc ,
(10.3)
The construction of the current vector is(t) is shown in Fig. 10.3 for assumed values of iS l > O, is 2, iss < O. Magnitude and angle each vary with time according to
The current vector determines the instantaneous magnitude and angular po sition of the peak of the sinusoidally distributed ampereturns wave produced by the three spatially displaced stator windings. By combining Eqs. (1004, 10.7) the ampereturns rnay be expressed as a travelling wave, the peak of which follows the angle a = ((t) of the current vector, 8s(a , t) I
r t1w
. I)
,I,
N :.'
1':1 (1 ),'
I"
I
/7,(/ ) " I"
1
I
I
f
~"
I
1
( 10.,1)
cos(((t) - a) .
(10.8) three phase and velocity
d ( / di, i II I,) I( ~ ai q~ap of the mo(;or . Siw(' 1.111' CC>lllp)(' X v('d()r~; ddillt'd iu )';(p :. (10.;, In.S) rJ('scribe t,he spat.i:tl dl ll l.r'il'lll,i ,," (Ir II 1I ,1I 11 ~ III'I , iI' lidd ill II, 1'""1<' Iwrp"lIdi('lrlll.r I,,, 1.1". 11101,01' Il.li i : l ,
(v I () ' i (II
= Nsis(t)
st a t.or currents are sinusoidal and form a syrnmetrical LII(! alllpprct.lIfl1s wave revolves with constant magnitude
s'ysL(~1I1 I
J't·lI.r,(,H III·. ir'l ~ . 1';'1. (I().:~) 1>('('OJll(~S
(10.7)
is(t) = is(t) ej(t) .
10.1 Mathematical Model of a General AC Machine
lU. SyJlllll(~Lrical Three-Phase AC Machines
1()8
they are also called space-vectors or space-phasors [N7, 873, 31]. However, they are not to be confused with the constant complex phasors for describing steady state sinusoidal alternating quantities. The sarne reasoning as before holds for the ampereturns wave produced IJy the moving three-phase rotor windings,
0 R (!3, t)
= NR [iRl(t)
cosf3
+ i R2 (t)
cos(f3 - ,)
+ iR3(t)
B( 0.)
cos(f3 - 2,)] . (10.9)
Incremental number of turns
N 2
.2. cos
WILcn defining a rotor current vector,
in(t) = iRl (t)
+ iR2(t) eh + i R3 (t) ei 2, = iR(t) ei ç(t) ,
169
Â. LLA..
(10.10) Fig. 10.4. Distributed winding
fR(t) = iRl (t)
+ iR2 (t) e-h + iR3 (t) e-i 2,
= iR(t) e-H(t) ,
(10.11)
(.]I(~
ampereturns-wave, excited by the rotor currents and moving with the ro(.or, assumes the form en(f3, t) = ~ N R [iR(t) e- i {3
+ i~(t) ei {3]
,
i
~s, ~ ~
(10.12)
real;
(t)
Ns
i Le; drcct. ou the stator is obtained by substituting
li -- n
() II
(10.13)
- é )
(II', (, I.) -
t NR [iR(t) e-i (a-c) + i~(t) ei (a-c)]
.
(10.14)
TI,,· ",: :"ll.;ut!, waw is a superposition of the stator- and rotor-ampere turns
H(
= 0s(0, t) + 0 R (0,
é,
t) .
= 2h 110 [(1 + 0"5)0 5 (0, t) + 0R(0, é, t)],
'l/J51(t)
= ~~ ~ (1+0"5)110
(10.15)
2
[e j À+ e-j À] /
/ ~~
+ N5.~~lr 110 /
[i5(t) e-ia
+ is(t) ei a] dodÀ
À-~
~
À+~
[ejÀ +e- iÀ ] /
_~
[iR(t) e- j
(a-c)
+i~(t)ei(a-c)]
dodÀ.
À-~
(10.18) (10.16)
il.CCOllnts for magnetic leakage; 110 is the perrncability constant. A lIloddlillg of leakage effects is not possibk with lhe simplifications
wli('I(~ (rs
""':til(~d
(10.17)
.x+~
1!. 2
2
rOJ'(:e becomes effective at the two airgap crossings, causing a local dClIsit.y at the stator side of the airgap,
/I,.;(n, s, t)
I r Bs(a, c, t) da] d>',
°
1.01l10!.i vC
1
,L '"'>.l2:
where I is the effective axiallength and r the radius of the rotor. The integra tion over is a consequence of the non-uniform field in the airgap while the integration ove r À is required by the nonuniform distribution of the winding. Inserting Eqs. (10.4, 10.14, 10.16) results in
,';illC" I.IL<~ pcrllleability of the iron is assumed infinite, the combined magne flIIX
i.
in the interval < À < Thus the ftux linkage in stator winding 1 is obtained by a double integration,
The evaluation of the integral is greatly simplified by the complex notation
Jllad('.
.
(1
WIL< ~ il' calcll];.d,illg llllx
lillkagcs, the sp;l.t.i;t! di::I,ribIILioll 01' thc conductors 1111 \1; (. IIC (·oll c. id(·J'('d ;t:; ill
N~ I r
+ 0"5) ~ 71' 110 N ,r.; NlIlr Hh
II.
-
71'
Ito
1
= "3 L 5 ,
=
:.li llli'I(' 1'<'11 1111. 'lfl Olol.II,ill('
1
M
:J .
,
(10.19) (10.20)
'l/JSl (t)
= %Ls [is(t) + is(t)] + %M
[iR(t) e j
171
10.1 Mathematical Model of a General AC Machine
10. Symmetrical Three-Phase AC Machines
170
ê + iR(t) e-j ê]
(10.21)
'file fiux linkage in the stator winding 1 is a physical integral quantity and Illust, of course, be real. The flux in the other stator windings is computed lilcewise, with the exception that the integration over À is shifted to , ± ~ and 2, ± ~ respectively, reflecting the position of the windings. This results
'l/JRl (t)
=
t LR [iR(t) + iR(t)] + t M
[is(t) e-j ê + is(t) e
j
ê] ,
(10.28)
which has a similar form as Eq. (10.21). However, as this fillX is defined in rotor coordinates, the stator current vector is now turned backwards relative to the position of the rotor. 2
(1
+ O'R) N;
lr
h
7r
/.Lo =
t LR
(10.29)
111
'l/JS2(t) = %Ls [is(t) e-h
defines the self inductance of a rotor winding. Likewise the flllX linkages of the other rotor windings are obtained,
+ is(t) eh]
+%M [iR(t)ej(ê-')') + iR(t)e- j (ê-')')] 'tPS3(t) = %Ls [is(t) e- j2 ')'
,
t
(10.22)
'l/JR2(t) = LR [iR(t) e-h
+tM[is (t)e-j(ê+')')+is(t)ei(ê+')')] ,
+ is(t) e j2 ')']
+% 111 (iR(t) ej (ê-2')') + iR(t) e-j (ê-2')')]
(10.23)
'l/JR3(t) = %LR [iR(t) e-
+t M
Tlw symmetry of these equations gives rise to the definition of a complex
v('dor of fiux linkages
(10.24)
=
+ M iR(t) e j
(10.25)
ê(t) .
','his IIIIX: vector describes the magnitude and angular position ofthe peak of 1.1((' :-; illllsoidal flllX density in the airgap ofthe machine. The exponential term aIJ;wllcd L() in (t) indicates that the rotor current vector must be rotated by
()r IIlcchanical rotation
COllversion of the stator current vector into rotor coordinates by Eq. (10.13) yields
("~:; ((J, c, t)
= t Ns
[is(t) e-j (.B+ê)
+ is(t) e
j
(.B+ ê )]
(10.26)
wilich produccs a fiux density on the rotor surface, corresponding to Eq. (IO,IIi)
/I}{(!1, c, t)
r' li
= 21h /.Lo [(1 + O'R)8R(!3, t) + 8 s (fJ , c, t)]
(10.27)
ilf'.ajll si.a.llds ror (",he tmavoidahle nmglldic 1r·;i1q~( ' . "d' ('I~raLioll aWlIlld t, h(~ ( ~ iI'Clllllf(~rclI(,(' (lI' ti", rul,ol', II.gll.iu prcsnpposing a
lil,l"ly d i ~ tri"IlI.('(ll'()tor WilldiJlI!', kad s to 1111 ( ' ~ pl' (' f1:il o ll fo r til(' JlIlX lillkitl~(~ iII
ro t"r
will"il" '. I ,
(10.32)
•
10.5. iR1
iS1
before its effect can be superimposed upon
~:;UII(' w;ty.
ê
The magnetic linkages described by Eqs. (10.21)-(10.23) are now used to derive the voltage equations for the stator and rotor circuits, depicted in Fig.
Rs
01' L1w stator current vector is(t). '1'11(' IIllx linkage of the moving rotor windings is computed in exactly the
or
(10.31 )
+ 'l/J R2 (t) eh + 'l/J R3 (t) ej 2')' L R iR(t) + M is(t) e-j
I.hal.
IIl!'illI~;
+ is(t)e j k+ 2 ')')]
[is(t)e- j (ê+2')')
!fR (t) = 'l/JRl (t)
I.cJ'lIlS
11.1'1,;1('
j2
+ iR(t) e ')']
')'
rotor flux .
allowing Eqs. (10.21)-(10.23) to be combined and disposing of the conjugate
(.11"
j2
(10.30)
These expressions are again simplified by forming a complex vector for the
!fs(t) = 'l/JSl(t) + 'l/JS2(t) eh + 'l/JS3(t) e j2 ')' ,
'~:.'Jt) = Ls is(t)
+ iR(t) eh]
l R, Rs
usd
u~
iS2
'l'SI
'l'R)
'l'S2
'l'R2
iS3 'l'S3
I
I
'l'R3
iR2 iR3
RR RR
I uR)
RR ["R' UR3
US3
_________________
~
L ________________ _
Fig. 10.5. Magnetic linkages and voltages
The tine-to-neutral voltages in the stator circuit are
. d'l/JSl RSZ81 + ----;[t = USl(t) U:.; 'i.:; ',~ U ; ';" . 1
d'ljJS2 _ 'tL'i2(t) , (li. dlf' :;;1 ti,
1I ;::I(t.) ,
(10.33)
ln
lO. Symmetrical Three-Phase AC Machines
10.1 Mathematical Model of a General AC Machine
wlwre Rs is the stator resistance per phase and USl, US2, US3 are volt ages of a.,iJitrary waveforms. These equations may again be combined by introducing ("/II\ljJlex vectors. By formal definition of a voltage vector, eventhough there is 110 immediate physical interpretation as with the currents,
"Ils (t) = USl(t) + US2(t) eh + US3(t) e j2 'Y, ;tllcI
(10.34)
with Eqs. (10.5,10.10, 10.25), this results in
. d'IjJs . dis d .. "S'/..5 + =RSIs+Ls dt +Mdt(1ReJé)=:J!s(t).
dt
1';lIlploying the rules of differentiation, this equation assumes the form lls is
dis
di
·
+ L S dt + M dtR eJ + j é
.
w M iR eJ é
= :J!s (t) ,
(10.36)
w!tere w = dê/ dt is the angular velo city of the rotor. The two terms containing t!t(, rotor current may be interpreted as voltages due to mutual induction and r/ lI.ation, respectively. l1y inserting Eqs. (10.21-10.23) into Eq. (10.33) and considering Eq. (10.1) whic:h is due to the isolated neutral, it is found that the line-to-neutral volt ; tf~('s aI' the symmetrical machine are also balanced at any instant,
'/ISI (t)
+ US2(t) + US3(t)
=O .
With the help of Eqs. (10.10, 10.32) they are again combined to form a vectorial equation,
. RR IR
(10,38)
(10,39)
clashc:d lines connecting the neutraIs in Fig, 10,5 are redundant in reality. 11/:111:(' t!te voltage equations are
/.Iw
"
fluiu/. 11'1l11l ~ 1
+ -(l1/Jra ,- = uw(t) , ri/. 111/' /,' .1 ,(/
d. _ . dt (1s e J é)
= :J!R (t)
,
(10.41)
+ UR2(t) eh +uR3(t)e j2 'Y
(10.42)
B RS ((3,
é,
N:~o
t) =
[i s (t)e- j (!3+é)
+ is(t) e jUHé )]
.
(10.43)
The magnetic field produced by the rotor currents themselves does not gen erate any net tangential forces with the rotor currents since there is no reluc Lance torque with the uniform air-gap; this could be shown by carrying out t!te caIculations with B R instead of BRS. The current distribution aR((3, t) along the circumference of the rotor is ;t!so, with the assumptions made, sinusoidal. According to Fig. 10.6, it is cldined as the spatial derivative of the rotor ampereturns,
- ~ 8GR((3, t) __ . NR [. -j{3 _ .• j{3] aR ((3 , t) - 2 8(r (3) - J 4 r lR e lR e .
(10.44)
T!te tangential force df acting on an axial strip of width r d(3 of the rotor sur ,'aC(~ is the product of radial flux density and axial rotor current distribution d!
= -B RS ((3,
é,
t) aR(!3, t) 1r d(3 ,
(10.45)
w!tidl Ilpon integration along the circumference yields the electric torque in clin:ctioll of rotation,
dt
({c/I 11.'1.
di
may be impressed by an external voltage source. ln case of a cage rotor, the voltages are zero and the rotor circuit is internalIy short circuited. The vector differential equations (10.35, 10.41) describing the electromag netic interactions of the symmetrical AC motor in steady state and transient condition are now to be supplemented by equations for the electrical torque and the mechanical transients. On a real motor the tangential forces are act ing at the sides of the slots, where the magnetic field enters the iron. Since there are no slots in our simplified model with airgap windings, the torque is computed with the help of the tangential Lorentz- forces exerted on the ax ial current carrying conductors orthogonally crossed by the radial magnetic field. The component of the flux density on the rotor surface which is caused by the stator currents folIows from Eq. (10.27)
'l'1c(' S
Ic./I'/./li
.
di = RR IR + L R di'R + M
(10.37)
IIIOst motors the neutral point of the stator winding is not accessible, ;:0 J.ltaJ. oltly the terminal voltages US12 = USl - US2 etc. are available for /'XI,CTII:d IIS(\ while the line- to neutral volt ages are dependent quantities. '1'11 i:: COIT( :sponds to an actual situation, where the motor may in fact be L1 /·/)/III('CI.(·d . With the terminal volt ages U12 = USl - US2 etc. the voltage VI'/'LIlI' iiI Eq. (10 ,34) can also be written as
i lll (I) + iR2(t) + iR3(t) == O ;
d'IjJ R
where alI variables are defined in rotor coordinates. The vector of the rotor voltages,
()IJ
(/ :: (/) = 1t12(t) - U23(t) e-h,
+
:J!R(t) = URl(t) (10.35)
173
!.II(' ·UII~(i.) '1/ /n
,
(I) .
271'
(10.110) 'III'M{I.)
l'/IU /,'
,/:lZ ( Uu s (f'i , It , o
/.)
11.11((1, /.) d(1 .
( 10.46)
10. Symmetrical Three-Phase AC Machines
J 74
10.2 Induction Motor with Sinusoidal Volt ages
1nserting Eqs. (10.43, 10.44) results in 2".
mM(t) = -
M. /
6 'Ir J
[is e- j (t3+ ê )
+ i*
ej (t3+ ê )] [iR e-j 13 - iR e j 13] df3 ,
· R Sls
. j ê) + L s didts + L O dtd (1R e = 'Y<s (t ) ,
(10.50)
· R R1R
' + L R didtR + Lo dtd (ls e- j ê)
(10.51)
-s
o
J
(10.47)
=
'Y
(t ) ,
~~ = mM(t) -mdt) = í Lo 1m [is (iR ej ê)*]
dê
where M is the coefficient of mutual inductivity defined in Eq. (10.20). When integrating over the full circumference of the rotor, the exponential terms containing f3 canceI. Hence we find mM(t) = -M 3 'Ir
/2". is iR e-j 2 ê
fSiR e J
-
o
j
.
ó
df3
= ~3 MIm [is(t) (iR ej Ó)*] (10.48)
the imaginary part of the bracket is equivalent to a vector-product, being proportional to the product of the two current vectors and the sine of the angular displacement X, as seen in Fig. 10.6 b.
dr
aR(f3,t)
dt
- mL (ê, w, t),
(10.52) (10.53)
= w.
Since the first two equations can be split into real and imaginary parts, this represents a set of 6 scalar nonlinear differential equations. They are valid for any waveforms of voltages, currents and at varying load torque and speed. The physical currents, for example in the stator windings, are easily de rived from the vectorial representation. With Eq. (10.1), the current vector, defined in Eq. (10.5), is
~ is1(t) + j
is(t) =
V;
[is 2 (t) - is 3 (t)]
(10.54)
hence (10.55)
is1(t) = íRe[is(t)] ,
BRs (j3,t)
175
and correspondingly, is 2 (t) =
is
í Re [is e-h] , etc.
(10.56)
Likewise the currents in the rotor winding are obtainedi of course, in case of a squirrel cage rotor, the notion of phase currents loses its physical meaning,
~ e ic
a
b
Fig. 10.6. Rotor current distribution and torque
When assuming that the stator and rotor windings have an equal number of turns, N~ = N s, and introducing the usual leakage factors,
Ls = (1 the
+ O"s) Lo,
M;;-:Lo
compld<' matlwntatical model of Lhe sYUIJllI'Lrjeal
with til(' llllllJwd illNI.ia J t.lw
L R =(1+O"R)L o ,
1Id.
10;1.<1
1.01"qlll'
t.JWIl aSSlllll(~K
:d. I.h(~ I'ollplillf':
01'
1.11(' followilll': !.II(' 1I1C/!.IH,
(10.49)
I~ellera.l fOl'lll,
AC machine mI, being
wil.h
lwen though the simplified theory remains applicable. The Eqs. (10.50-10.53) are the starting point for ali subsequent discus sions of transient and steady state phenomena. They describe the electrome chanical interactions of a symmetrical AC machine and can be adapted to various constraints, imposed by the type of machine used, as long as the basic assllmptions on which the model was based, remain valido This mainly refers 1.0 ra.dial symmetry and constant airgap, excluding saliency of the stator or rotor. Table 10.1 contains the rotor circuits of different nonsalient AC machines IIs('d in controlled drives that may be represented by the mo dei equations. 'l'hc details of the constraints are discussed in later chapters. 'file model equations contain, of course, also the special case of steady :: !.al.e opcration, when the AC motor is fed with sinusoidal impressed volt ages 1J)' a s'yulluetrica.l three-phase supply and is loaded with constant torque. To f01"l1l a link to tbis commonly used presentation and to gain a better IIlId('l"st.alldilll ~ of \.III: 1llOdel <'quntiolls, t,his cas!! will bc discussed first. With 1111
0,
H
( 'I ~I \"
IIIO!.Ol"
IO!.OI' 1' .' : Ii: I!."I' :: III!!}
or a wC/ulld rot.ol'
Itn,v,' 1""'11 ,1.11<11 '01 .
1'1 'S II 11.:-1 ,
w!tI'J"(' ('xLern a) ~ymmdrical
,'(1;
'II . ~-; y "'lfIetrical
Three-Phase AC Machines
10.2 Induction Motor in Steady State
'1'11 101. · 10.1. Mathematical models for nonsalient rotors
,,' di'I'"I" '''1. AC machines used in controlled electrical drives
.
iR
IJI R1 iR1
= '23/F
iRI
US2(t) = USl
i'RI
RR
US3(t)
IJI R2
.
(t - :J
-/2 [U e j 2 -s
.,
+lR
= USl
t-,)
(Wl
+ U·
-s
e-j
(WI
,= ""3 21l'
t-,)]
(10.58)
(t - ~~)
-/2 [U e j 2 -s
3 /
177
t-2,)
(WI
+ U·
-s
e-j
(Wl
t-2 ,)]
(10.59)
lR='2 F
IJI R3 iR3
a Synchronous motor with permanent magnets on rotor surta ce
~ !dR
b Synchronous motor with permanent magnets and damper winding
~IUF
~
c
The voltage phasor U S is a complex constant, having a magnitude equal to the RMS line-to-neutral voltage and an arbitrary phase angle TI which may be set at will because it is defined by the chosen instant for time zero. When the complex voltage vector according to Eq. (10.34) is formed, a very simple result is obtained,
=const.
iRI
~
= O
!dR=
Induction motor with wound ar cage rotor
URI
i R2
UF iR3
d Synchronous motor
wíth fie/d and damper windings
e Doub/y-fed inductíon motor
. . () = USl + US2 e J , + US3 e J 2,
Jfs t
3V2
.
= -2- -s U e J Wl
t .
,
(10.60)
the conjugate complex part vanishes, leaving a vector with constant mag nitude and constant angular velocity describing a circular path around the origino The stator and rotor currents form also symmetrical three-phase systems in steady state. With
with impressed rotor vo/tages
is 1 (t)
i (t) = 3 -/2 I e j -s 2 -s
IO.:l . l Stator Current, Current Locus
II. :i'y lIlllletrical three-phase system of sinusoidal voltages having the angular
'll.S J
(t) =
-/2 uS
COS(Wl
t
+ Td
= Re
[-/2 Us
= V2 [u ejwlt+U*e-jWlt] 2 -8 - s , w IIC:re rro.; = (Js
(;j T I
TI
ej
Wl
i
t]
t
(10.62)
,
RI
(t) =
-/2 2
[I-R e j
(WI
-w)
t+ r
-R
e-j (WI-W)
t]
,
etc.
(10.63)
n" (10.57)
is t.hc constant volLagc I'IJ:tJ;or; correspomlinG exprcssions
ii .. !" for I'lItuws ~ 1IJl
WI
where the current phasor Is is still to be determined. A similar definition ('xists for the rotor currents oscillating with slip frequency W2 = Wl - W
may be defined as follows, using complex notation,
ej
(10.61)
we find
10.2 l.lHluction Motor with Sinusoidal Symmetrical V olta/,!;es in Steady State
I'r(:<jIH:JlC:y Wl
= -/2 [I e jw1t + -s r e- jw1t ] , etc. 2-S
. . _ 3 -/2 I ej(WI-W)t. L/I(t) 2 -R
(10.64)
TI\(: rotor Cllrrcnt in stator coordinates, i.e. as seen by a stationary observer, III with ~ - wl. lWC;lUSC of constant sp(:ed, i /{(I)
, ,, I tI {l}
., ..
J/. '0
I/{
i (",I
t ..
(IO.(;[i)
I '/ H
lO
''-;'y"11I<'l.rical
Three-Phase AC Machines
10.2 Induction Motor in Steady State
II(~II(,( ' Ilw rotor ampereturns wave, moving with slip frequency across the rotor slIfface, rotates in synchronism with the stator current wave; this is a
prccondition for constant electrical torque. By introducing these expressions into Eqs. (10.50, 10.51) and omitting the redundant terms, the differential equations are converted with UR = O to algebraic equations for the phasors,
+ j Wl as LoH s + j W} Lo (Is + IR) = U s (RR + j w2 aR LoH R + j W2 Lo (Is + IR) = O.
(Rs
N orlllalising
(10.66)
(10.67)
Wl
n'snlts in (
~R + j Wl aR LO)
+j
IR
Wl Lo (Is
+ IR) = O.
(10.68)
I~:qu ations (10.66, 10.68) lead immediately to the familiar single phase equiv alcllL circuit seen in Fig. 10.7 a; its validity is restricted to steady state, asslllning sinusoidal symmetrical currents and constant load torque.
as Lo
/Is
aR Lo
lR
aLs
RS ~.
J.s 1/.,'
LO
Is
1.1/1
Y.s
S
(1-aJ Ls ,
=.1m
..!!....L 1-a s
1-/S
1 RR --2 (1+aRJ S
lmR
Ns =N'R
;/
(1+aRJ1R
is RR
I
b 1
--lR 1+as i
II
is
y.s
Lsl
O(1+asJ SRR 2
1ms
·SWII7L R
1+J RR Zs=Rs+jw}Ls l+js'1: R
where
Sp
W}
RR a LR
1 wlaTR' with
F iH· 10.7. Silll\k pI IlL~(' 1.''1"ivaklll. c.ircllit,j "I' i,,
'"I
'"I jl!l
Il llLl,,,'
,,,"1 1'0(.,,,
:: I,at" .. Ili
1'0(,01'
111.1, ' ,
LR TR = RR '
(10.71)
which is called pull-out slip because a motor with zero stator resistance pro duces maximum torque at S = Sp, as will be shown later. The pull-out slip of normal medium size motors is Sp < 0.25. The steady state equivalent circuit in Fig. 10.7 a contains three synthetic inductances which, because of 1m = Is+IR' are not representing independent energy storages; hence it must be possible to redraw the circuit diagram with only two inductances. One possibility is shown in Fig. 10.7 b, where all the lea.kage is concentrated at the stator side. There is a different fictitious rotor resistance leading to a different current in the right hand side branch of the circuit but the stator impedances are identical. The new synthetic rnagnetising current ImR represents the rotor flux, as will be shown in a later chapter. A third equivalent circuit is seen in Fig. 10.7 c, where all the leakage is referred to the rotor side and where Ims stands for the stator flux. This arrangement can also be used for designing controls. The stator current is now computed with the added simplification Rs = O which is of little significance with larger machines operating with low slip at line frequency. Thus the normalised expression for the stator current phasor becomes I -s
(h) wil.h 1"111 (1 11 1'" ;,,<111<'1. 111,11'" (. -) wi!.11 1(l1I 10 I/ J.4' IlIdllc (, 11 1" '1\
=
. 1
c
(10.70)
+ as)(l + aR)
is the totalleakage factor of the motor, which has an important influence on the characteristics of the machine; it is chosen by suitable design parameters such as the shape of the slots and the length of the airgap. Normal values of a range between 0.05 for low leakage machines to about 0.20. The expression in Eq. (10.69) gives rise to the definition of yet another important parameter
1+as
(II) w;l.h
(10.69)
'
1
s = W2 = W} -w Wl
By inspecting the equivalent circuit in Fig. 10.7 a, the stator impedance per phase is derived. It may be written in the form
a = 1 - (1
the second equation with the slip frequency
179
=
S
U 1 + J Ü s. -s ,s j Wl Ls 1 + J sp
(10.72)
whj rJt is n liucar cowplex function umpping the real S/Sp-axis onto the Is pIHII( ~. A('('ordilll~, to t.Iw mies or confoJ'lllal Illapping, the result is a circular lorwl, 1.111' ('irei" di:"',1 i l,tIl , ...lS(l ('1I.lI('d ( h rmllll l l.'H (lI' I!('ylaud's circle . This is ""lIlirll"'.! 111'y LI", 1,')It>\v ll,,~ I.r; ~ II:lr(lrlll l l, I , i ,,,,
180
10. SymmetricaI Three-Phase AC Machines
Is = Iso [ 1 + O' 2 O'
10.2 Induction Motor in Steady State
1 - O' 1 - ~ ~l 2 O' 1 + J sp
_
= Iso [ 1+0' _1-0' e-j2arctans/ sp], 20' 20'
(10.73)
where
representing (for Rs = O) the no-load magnetising current as mentioned before. Minimum phase shift cp = arg(U 5,15), i.e. maximum power factor is reached when the current phasor forms a tangent to the circle. The optimal power factor then follows from the geometrical relations of Fig. 10.8, COS(()
rn
Iso = j
Il. s Wl
(10.74)
Ls
is the ideal no load current valid for S = O, i.e. when the rotor moves syn chronously with the revolving stator field. When no speed difference exists, no rotor current will be induced and there is no torque; hence the no-load current is a purely reactive magnetising current. This is in agreement with the equivalent circuits in Fig. 10.7 for S = O.
(10.75)
'
7r
cp = - -
2
1 S S - - arctanSp Sp
+ arctan -
(10.76)
O'
by differentiation , d(trSp) = O; the result is
VU.
=
(10.77)
p
Rence, not only the geometry of the current locus, but also the maximum power factor, the pull-out slip and the normalised slip at the nominal oper ating point are all determined by the leakage factor 0'. The magnitude of the normalised stator current,
.I.
!,Is
S 1 . / " Sp
,
~
1- cr e - j2 are ta{ Sp
/
cr JSD
Is Iso
(ínductíve)
-1
Fig. 10.8. Locus of stator current with stator resistance neglected, circIe diagram for an induction motor
is an even function of slip, as seen in Fig. 10.9 a. The normal operating region comprises only the central portion of the curve. At larger values of slip, for instance when starting the motor at S = 1, the current rises sharply so that the simplifications employed become questionable; this is particlllarly true for the iron saturation that is not negligible at high currents. The stator current at nominal slip, Sn/Sp = y'ci, is
IS n
Iso The complex locus Is (S/ Sp) is depicted in Fig. 10.8 with the voltage phasor assumed imaginary, U 5 = jUs, in order to place th e circle in the customary orientation. The curve covers the fllll rall~C "00 < S/Sp < 00; for S > O the motor runs below synchrolli~IIJ , drawillg acl.ive power, while for S < O iI, op( ~ ral.(~s as a gcnerat,ol' alld Sl1ppli('/ol 1\cl.iv( ~ pow(~ r to t1w lilln. Reactiv( ~ pownl' ror lllil(f,lwti satioll is il.lwaY ~1 11(.,..11'11 whi('h ,i!l n dmra('.I.(~ri!;tic of illdllt'l.ioll 1I 1I\f IIi Il ('1i Jlil.vil lj ( 110 illl."l'IlII.J v,,JI.II II.'·
til. S
II 1.1".
11 11111'("( '
whi('h ir, pn ·r;(·IJI. iII
1'1 111"' 111. HIi!! " " " '!1 il. rl II d llillllllll vILIII"
(10.78)
VI + (ir
Reactive current
cr 2 cr
, 1+ JSD
V1+(~ff;r
1
- Jso 2 cr
JSD
HY III·h,.IIl1 ll ll il Il lhe lllll('" .
1-0' 1 + O'
= -_.
this is usually defined as the rated or nominal operating point. The pertinent value of Sn/ Sp is obtained from the phase angle, Eq. (10.72),
~n Active current
(motoring)
Js
181
1
= y'ci
(10. 7!l)
,
while the asymptotic value of the current is 1
Is Iso
fim 5-400
(lO.BO)
O'
Tllis ull(krlillCS I.he importance of lhe feakage factor O' as a design 'I'lw III<\ Xillllllll ;tJl(.l ruilLimulU val!i(':; "f til" ac:t.ive current pO\V('r, ;1,S;; lIl1lilll'. Ii..: ('OIlSt.., ;tJ'(' 1'1'11..-111'11 H,1. S (lI' 1,111' .. ird,·. I" LIli' III' XI. ró( ·(·I.j"lI íl. 1:1 !.h"wlI I.h:d.
wl"'I"
1.111' 1."1
ii""
11 : ./1 111111' 11 iI ,1I
P('It!( vrdll'·:.
pararnd( ~ r. anel
acl.i v,~
1''''1'' i.". in tlw v( ~ rl.i( : (·n
LlI<':'w a.re
al ~ ()
1.11(' poilll.:.1
10.2 Induction Motor in Steady State
10. Symmetrical Three-Phase AC Machines
182
3 1 - (J U~ mM = - - - - - - -
~
2
ISO ~~--------------I1 - ---------------;~
--',
1 ,---
//;
RS=O (J
=0,1
ISO I
I
-2
-1
a
Sp
S
2
-~ ~
mM = mpo
Sp
3 1-
úJ,mM\
1-----
mn mpo
I
p
.... _ _
I
: Motor
s
úJ 1 _ U úJo - Uso
=1
I I
I
-1
S
2
Sp
Generator
------------' -1
Fig. 10.9. Stator current and torque as a function of slip (Rs
= O)
10.2.2 Steady State Torque, Efliciency
An expression for the steady state torque is obtained by inserting the vec tors for sinusoidal symmetrical currents, Eqs. (10.62, 10.65), into the general formula for the torque, Eq. (10.48), mM = 3 Lo 1m (Is
IR) ,
(10.81)
yielding a constant value, as expected. From Fig. 10.7 a the following relation between stator- and rotor-current is derived, -j w} Lo I I -R= RR I S+jw} L R -s,
(10.82)
which leads to mM
'
:\ r ' lI
Uslw} Usolwo )
2
2
SISp + SpIS '
(J
U~o
SISp + SpIS '
(10.84)
(10.85)
(10.86)
mo=-------
mpo
b
2
=m p - - -
where
mM
-2
(
Sp
,
I
2
SISp + SpIS
where mp represents the peak torque, available at the pull-out slip S = ±Sp. The function of torque vs, normalised slip is plotted in Fig. 10.9 b. When the induction motor is part of a variable speed drive, the stator voltage Us and frequency w} may differ from their nominal values Uso, Wo shown on the nameplate; hence the following extension of Eq. (10.84) applies
/
~
w? Ls
(J
183
j
l ,~; TIll [ ali I S
wl
I,o
.i W I
Wil.li 1.111' 111'11' "I 1-:'1 , (I (J ,'tH) I,liil :
[,1/
(' 11 11
I.
1" ,
)"',111 11 111',""
(10.8:3) L..
2
(J
w5 Ls
is the maximum torque (for Rs = O) at nominal volt age and frequency. This indicates that the torque rises with the square of the flux; the reason is of course that the rotor current, being caused by magnetic induction, is also depending on fluxo The torque vs. slip curve seen in Fig. 10.9 b may be qualitatively inter preted as follows: When the rotor speed deviates from the angular velo city w} of the stator field, currents are induced in the rotor winding producing a torque with the tendency of reducing the speed difference. Below synchronism (S > O) the torque acts in the direction of the rotating field, i.e. the machine operates be tween standstill and synchronous speed as motor; above synchronous speed, the torque is opposed to the direction of rotation, hence the machine acts as generator, feeding electrical power to the line. Steady state supersynchronous speed can thus only occur if the machine is mechanically driven . Because of the reactive power problem, induction generators are rarely used except in small unattended hydro or wind power stations, but temporary regeneration is a normal feature of controlled AC drives. The occurrence of a peak torque can be explained by a simplified model of the rotor circuit as shown in Fig. 10.10. The rotor may be represented by a R-L-circuit that is supplied from a voltage source, the amplitude U2 of which is proportional to frequency W2. The current phasor
U20 ~ Wo I 2 = R 2 + j W2 L 2
U 1
-20- - jwOL2 1-j ~ W2 L2
(10,87)
as a function of frequency W2 is described by the semicircle shown in Fig. 10.10 b , While the magnitude h rises asymptotically with W2, the active componenl. H,(' / '.1
(}~:o ii.!"
I J'.~
1 II ~ 101 ..
1, ..1
ti/ :l l i ~
/I "
(IO,RR)
184
10.2 Induction Motor in Steady State
10. Symmetrica! Three-Phase AC Machines
J2
U20
L2
Re
'2
W 2 W
a
b
o
j WOL2
~
I I I I
Motor
0<S<1
C
(10.89)
'
wi Ih , ~xt:ernal resistors in the rotor circuit then offers a possibility of modifying lhe torqllc/speed characteristics by changing the pull-out slip Sp through the rcsist.ance. The resulting family of curves is depicted in Fig. 10.ll in the
;\,n :lls toIllcd orientation. ln combination with the dashed curves for inverse
of the field by interchanging two stator t erminais a fairly complete of the four quadrants in the torque/speed plane is achieved, albeit 1.\1(' sliape of SOlllC eurV($ is unfavourable. Lowcrilll-!; the st.at.or voltages, for iIl S t,"'Il(,( ~ I.hwll f.': 1t a tap cltanl'-(illg trans[oollcr, l"(~dIlCCI ; Lhe peak torqllc by the ::(\,,<1.1'" 01' l,b <' voll.a!';,' \'(~dIlCl.i()lI, f\lrt.ll<~r ('olllpr('::: :illl'; 1.11(' (",IlJ'V(~S itt bori7.Olll.al
H\ l tI,1i 1'
01' (·Olll.íIlIIOIl ! : 11 ( 1(" ' "
'·"lI l. lt" I.y
rol.o!'
/ / /
/
time duty, for instance when starting the motor or with crane hoist appli cations, are these los ses tolerablej sim pie control schemes of this type are in widespread use. The efficiency of the simplified symmetrically fed induction machine in steady state operation may be assessed as follows: Because of the neglected iron and stator copper losses the active power Ps supplied to the stator windings is passed on to the rotor, where it is transformed into mechanical power and heat losses. ln the steady state equivalent circuit (Fig. 10.7 a) alI this power is dissipated in the synthetic resistance RR/S, P
R
= 31 2 RR
(10.90)
RS'
ln reality the resistance in each rotor phase is only RR, hence the difference in power must correspond to mechanical power that cannot be modelled directly by a static circuit, Fig. 10.12,
Pm
[1 ]
l-S · = 31R2 RR S -1 = 31R2 RR -S-
(10.91)
This yields an estimate for the efficiency in the motoring region, O < S < 1, 1'", 'II",
}"( 'Sí HI.O!'K (lI'
vnl i'.HI'." t' b UI!'
-2
Fig, 10.11. Torquejspeed curves of a wound rotor induction motor
I" '(' I. LOII,
dI/tiNI ,",(,1,
S>1
/
phase sequence
"í
IlIn i ll
Reverse braking / / ----
with reverse
rol.aÜOll
l 'q l\I"III'i', til"
-1
Operation
wvnag<~
'1'11('
mM
mpO
--/-------
11I;\,xiUIIIIll torque but only influences the pull-out slip. A wound rotor motor
I',V
/
(:~-/S~=0,2
whidl is a measure for the overload capacity of the motor. The pull-out I"rqllv , Eq, (10.86), shows that the rotor resistance has no effect on the
rot.or
-0,5 ~/
-1
I"
1 + (7 2 fo
Generator
.R2
passes through a peak value; the curve, plotted in Fig. 10.10 c, has a similar shape as the torque characteristic in Fig. 10.9 b. This points to the reason for the decreasing torque of the induction mo tor ato higher slip frequency: Due to the rotor leakage, the rotor current is increasingly lagging, causing the rotor current wave to be shifted and eventu ally become orthogonal to the stator flux wave. Hence the torque is reduced whereas the stator and rotor currents continue to increase asymptotically. TIl view of the stability of the drive and the rising losses, steady state opnat.ion of the motor makes only sense below pull-out slip. WhCll declaring the point, where the power factor is optimal, as the nom ill:!.1 op,'rat.ing point, Sn/Sp = fo, the ratio of the peak to nominal torque
-
, RS= O
USO
w2 L 2
F i g. 10.10. Simplified mode! of rotor circuito (a) Circuitj (h) Locus of current phasorj ( c) Real current component
_
o
S
U20
'1I1' JI
W
W1
2
U20~O
'111'1.0
!:!L= Us
W
185
H.:;
{'II
W
(10.92)
I - S = -< l , WI
w..I1 11f1 /'", "1" "'11.11111 : I :~ ;~ 1 ~' · "(·l' n, I,,, I' ! .'í
()
186
10. Symmetrical Three-Phase AC Machines
PR _ _1_ =
rJg = Pm - 1 - S
Wl
w
10.2 Induction Motor in Steady State
<1
(10.93)
which means that the efficiency even of an idealised motor ü; lowered in proportion to speed, whether the reduction in speed is caused by additional rotor resistors or by reduced stator voltage. A similar situation exists in the generation region, where the losses rise also with the magnitude of the slip. With a real motor, having primary copper losses as well as iron and friction losses, the efficiency will drop below this idealised case.
iR
(JRLa
PR ~
Inputpower
RR
RR
S
~
Pv
Real
Model Rea1
for a = 0.1,
no = 750 l/min
--s;
fo
= 0.22
0.20
0.32
0.30
fs!
fo
= 0.22
0.30
0.32
0.40
cos 'Pn
l-a rra = 0,90
0.90
0.82
0.84
mpo mn
1~ 2 a
2.30
1.82
2.0
Sn
Pm ~ Mech. power
The explanation is, of course, that the induction motor functions like a mechanical clutch, where the product of speed difference and torque causes an unavoidable power loss. The definition of efficiency in the braking region at reverse speed, S > I, does not make sense, because electrical as well as mechanical power is supplied to the rotor and converted to heat. With high slip control schemes for short duty drives the main problem is the dissipation of the rotor heat loss. ln this respect a wound rotor motor with external rotor resistors is preferable, however at the cost of moving contacts. These results give ample evidence that an efficient adjustable speed AC drive can only be realised when either the slip power is recovered by some means or the stator frequency Wl is changed to track the mechanical speed, thus restricting the rotor frequency to small values. 10.2.3 Comparison with Practical Motor Designs
The simplified theory of a symmetrical induction motor in steady state points to the totalleakage factor (j as the most inflllcntial panl.111 ctcr of the machine. As was mentioned before, (j depends on thc :-;11,,1'(: or til!: Klots and increases with lhe airgap. Tlw seal,' or til ,: círc nlar curwlll lonl~; iH d('I ,"l'Jllill"" lIy Uw i(!.:a,] Ilo.. load ('111'1'('111 , l .,." wh i(' h lI hw "",)/,11,11-1 1.0 a la r/ :" ,' x!.,' .. t "II 1,ltI' II i 1'1','a." , 1",,1' 11I1'c!uwi(';d 1" ' /l i,I II " ',
Model for a = 0.05, no = 3000 l/min
)
Fig. 10.12. Power balance in the rotor
'1 1'1,(1
The scale of the normalised slip/torque curve, Fig. 10.9 b, is fixed by the pull-out slip Sp and the peak torque m p ; in the interest of good efficiency in steady state, Sp should be as small as possible. The peak torque is also affected by the leakage factor. ln Table 10.2 some of the characteristic parameters obtained with the simplified model are compared to those of standard motors, as taken from manufacturers lists on medium size (> 100 kW) 50 Hz motors, for a two pole and an eight pole-machine. The comparison indicates a fairly close agreement, at least from the standpoint of the user. Table 10.2. Comparison of idealised model with pararneters of standard motors
Rotor lasses
,, ,, 1 RR(S-l)
,'r'
1,111' IlI r j ',!I'I' LI 1/1, 1'/ ',"1' wil,11
IlJlrll. l l",)"
/I lrll" Í, WII
187
1',,11' 1II;I('I,illl': ; (Ir "qllit.l
= 2.35
10.2.4 Starting of the Induction Motor
When selecting suitable values for efficiency, power factor and overload capac ity at the nominal operating point, the starting performance of the motor is unsatisfactory; this is apparent from Fig. 10.13, where characteristic variables of a motor with a leakage factor (j = 0.05 are plotted against normalised slip. The conditions for acceptable performance at the nominal operating POill(. are:
a) The normalised slip at the nominal operating point should be about SnlSp ~ ..j(i to achieve a good power factor and adequate overload ca pacity_ b)The need for good efficiency rJn < 1 - Sn calls for a minimal value of nominal and, hence, pull-out slip. With motors above 100 kW, a nomill;d slip Sn ~ 0.02 is realisable; the pull-out slip would then be about SI' ~ O,lO. This llw:tns that an indllctioll motor with a short circuited, consta.lli f':" :iist,nIlC<' rotor wOllld llitvo to Ht.ari fnllll fi lalldstill (8 = 1) at SI Sp ~ 10 whirll c'o[' n 'R!JOlld N !.o n ll op e ra.l.ill ~: POilll, ",I, 1.111' I"i/',.
I O.H. ' 1'1((' il l,lIrl.ill/ \ 11I'1'f'mfll ll llc'l' ;11, S
rll':hl, 01' !.h, ~ circula,!' I , "!a n rll l'I,(' , j!1c·t1
filJ'
"y
dia,/~nulI ,
10.2 Induction Motor in Steady State
10. Symmetrical Three-Phase AC Machines
188
IS mM I 'mn
with Bp
COS <{>
Sn
Is
R==O
3
Snc
", ,
mn
2
-==-'0= 0224 Sp '
,,
l-O
COS <(>n =~
, '........ ........
== 0,9
mM
. . . . ---_'!!.n_ Nominal operating point
, 0,5
~
~ ~ 2.35, mn
0.66,
signifying a decisive improvement over the case with a short circuited ro tor winding. With larger motors the starting resistor may have the form of a tank of water with enhanced conductivity into which the electrodes connected to the sliprings are progressively submerged. b) A more elegant method requiring no moving contacts is the utilisation of eddy currents in the rotor bars leading to a concentration of the current in the outer portions of the conductors at higher rotor frequency (skin effect), thus increasing the effective rotor resistance at larger slip. As a consequence, the starting torque is improved with an accompanying reduc tion of current, while in normal operation at low slip the desirable features of the motor with short circuited rotor are maintained.
0= 0,05
~
coscp
Sn
I Sr
COS <{>n
= 1.0 would show the following starting performance at B = 1
- I ~ 3.2,
IS
4
2
3
4
5
Yci
189
S Sp
R2
Double cage
Fig. 10.13. Characteristic curves of induction motor with constant rotor resistance
-1 I
Is Sn
~
1
-lU
~
4.5 ,
cos cp
~
0.095 ,
~ ~ 0.47, mn
would be very poor, since low torque is combined with a high predominantly reactive current. When the motor is fed from a grid with a relatively large internal impedance, there could be an appreciable volt age drop, reducing the current linearly but the torque with the square of the voltage. This may even lead to a situation where the motor does not start under loadj the protective devices would then have to disconnect it from the line to prevent damage due to overheating. The conditions are particularly severe in small grids such as on board ships, where the generators are not by orders of magnitude larger than the largest motorsj this also calls for fast volt age control of the generators. Two options are available to improve the starting performance: a) Use of a motor with a wound instead of a cage rotor which permits the rotor resistance and, hence, the pull-out slip Bp to be temporarily increased at will, creating optimal conditions for starting as well as aI; operational speed. The rotor resistance may be shorted in steps by COllt.a.ctors controlled in a fixed time sequence or depending on stator CHr)'(!Il(. 01' spced (sec also Sect. 3.1.4). The larger part ofthe power losses dmil'l'. st.il.l't. o<:(',llrs ill Lhe startiug resistor, i.e. outside the motor, so UlilJ 1.11<'1'1' ;1,1\' 110 addit.iollal coolillg problC:lll~;, Wll(~u t.\l(~ operat.iIlI~ s()('(' d 11 11:1 (" 'V II 1'(· n.cll('d, i,ll(' slip-rin l(S call 1>c s!lol'l,,'d illt.l'I'l lII. lly wit.h ii ~ (I('('i ll. l 1111 '1' 111 11 11 11 111 wl li ll' t.l1(,'IH'II :d ll': i 11.1'11 lil'l,('" 1.(, !-tv" i ,1 LI,, ·, 1I1111< 1I'(' WII II' y f l'lt- Ll oll (lI ,,1 W(' 1'.I .
fi. 1""1., ,,· wl l. "
L2
(10.94)
/I
o,() r"
II I.I t l' l. I "I "
a
b
R I "R2
LI "L 2 Fig. 10.14. Cross section of eddy current rotors and equivalent circuit
There are various designs which tend to emphasise the effect of eddy currents in the rotor bars. Qne solution employing two cage windings is shown in Fig. 10.14 a. The starting cage (1), being of higher resistivity, is placed in open slots immediately below the rotor surface and exhibits low magnetic leakage; as a consequence this winding produces a high pull-out slip BpI. Because of the low leakage this winding dominates at high rotor frequency, i.e. at low speed. The operational rotor winding (2), in contrast, is characterised by a larger cross section of the conductors, possibly employing also material of higher conductivitYj the leakage reactance is increased because the bars are now embedded in the rotor iron. Clearly, this winding becomes mainly effective at low slip fre(]IlPIlcy, Figure 10.14 b shows the equivalent circuit with the two rotor Willdilli\S COllIH',c[,cd in parallcl. 1\';<:;1.11:';1' 01' II" n~, DI ~ L~ 1.111' I.wo willdiJlgs ar<~ cOlllplemclLtary. '1'11(' ('tlnl·1I1. 1",'11:: ,,(' !.II(' 1IIi1,dlilll: iH, "r ,' 111 II':!" , II<' 1011/':('1' ;1. ('irdc, 'I'ypic
1'111
I,. /
,'. 11
III.
111
..
,"iY"lIlldrical Three-Phase AC Machines
10.3 Induction Motor with Impressed Voltages
191
M
11111
~
,/
'Sn
?
mM
---------
mn
Should the motor be fed with sinusoidal symmetrical three-phase voltages of constant frequency and loaded with constant torque, the transients will eventually decay and the steady state condition discussed in Seet. 10.2 will emerge, where alI quantities in the stator are periodic funetions with the angular frequency Wl' A better insight into the dynamic behaviour of the machine is obtained by choosing a moving system of coordinates, where these steady state oseillations disappear. A suitable choice of coordinate system, independent of the waveform of the voltages, is that defined by the vector of stator voltages, Eqs. (10.34), (10.38),
'Y,S(t) = USl(t) + US2(t) eh + US3(t) ej2 --y
Jo'i(.;. 10.15.
(10.95)
1 S
Sr
= UI2(t) - U23(t) e-h = Us(t) ej
Characteristic curves or induction motor with eddy current rotor
À(t) ;
its instantaneous angular velocity ('llrves of an induction motor with double cage rotor are seen in Fig. 10.15; ;\ wmparison with Fig. 10.13 indicates the improvements achieved. Another d(~si~1I of a rotor winding with strong eddy-current effect is a cage with narrow I)ar,~ extending deep into the rotor iron; this is also seen in Fig. 10.14 a. WIt(~n line-starting larger motors it is sometimes necessary to reduce the ílllllSIt ClllTCnt by lowering the stator voltages or by using only part of the ,·;L:!i.(I/, willdillg, thus increasing the leakage factor. ln both cases the starting (,(lrqll(' is fllrLllCr reduced which limits this method to starts with light load. ' I 'II\' ,;i"'lll('sl. possibility for reducing the voltage during the start is, of course, " y 1/', 1'I '( '(lIIIJ{:ction of the stator windings. It entails a reduction of voltage 1\11<1 "111'1('11(, hy 1/../3 and of input power and torque by 1/3.
d>'
corresponds to the angular frequency of the stator voltages. Should the mo tor be supplied with sinusoidal symmetrical three-phase voltages of constant frequency and amplitude, Eq. (10.60) again becomes yalid
'Y,S (t) = 3 y'2 US ej
SlIpplied by Impressed Voltages of Arbitrary Waveforms Arl.,'!' I.his discussion of the steady state with sinusoidal symmetrical three voltages let us now return to the more general model of the symmetrical i11<1 Ilcl.iol1 motor in dynamic condition, when torque and speed are varying. Wc assume that the stator is supplied by impressed volt ages having, in prill('iplc:, variable frequency and arbitrary waveforlll , as long as Eq. (10.37) is rlllfillNI. lu fact, due to the isolated neutral point o[ the motor, even a viobl.ioll of thi::; eondition would be of 110 COllS(:qll('l1n:; iI. would simply mean I.ha!' I.he llel1tral of t:he WilldillgHaSSlIm(','; a dilf(:n:lIt pol.( 'lllial aoS the neutral "r 1.11(' slIpply vol(,ap;<'s '/1',' :1, '/},:;2, '1/.,'::1' AlHO, til(' H'HIlI!,!! IlIlly III' g'(:n('raliscd to rI. lill/,/,Iy h;wjl'l ~ ('()II~I.;l.Id, :tlld (:'111111 illl.('!'ll ld illlj"'t!HIII'\'H iII (';LeI. pha.;.;('; 1.I":y ('",ii.! I... ill<'llI,[(',1 iII U.., ,, 1.11.1."1' illlJII'd lu""' 1I (1,' ,'1 , ,, :: 1,(,) "r 1.111' 111"1.,,1',
(Wl t+Tl)
2
(10.97)
'
resulting in
us(t) '(I.;, I Hd Ilction Motor
(10.96)
Wl(t) = dt
3y'2
= -2- Us,
>.(t) = Wl t + TI ,
(10.98)
where U s is the RMS line-to-neutral voltage. All electrical variables are now defined in this coordinate system; i.e. they are transformed as they appear to a moving observer positioned at angle >.(t); by making use of Eqs. (10,7, 10.10) we define
= is(t) ej (-À) = lis(t)
is(t) e-j À
phasc
(10.99)
;l.Ild
iR(t) ej ê e- jÀ
= iR(t) ej (e+ê-À) = li R(t)
(10.100)
as the current vectors in moving coordinates. With sinusoidal symmetrical
I.hree-phase stator voltages of fixed frequency and constant load torque, these qll<1l1ti (,ies are constant in steady state, When multiplying Eq. (10.50) by e- j À w(~ JiJld ,
/(, " 1" . 1'
",
.1
\
Ih -;
I { ",. . ' .II.
t:
"À
J
1/'(1
d( 1./ "1'.1 , III 1
',\
)1 ' J
'Il.
i s r-· >--
(10,101 )
192
10. Symmetrical Three-Phase AC Machines
10.3 Induction Motor with Impressed Voltages
which is transformed by substituting
(10.102)
Developing the expression dI' · dis -jÀ iS -e =--
dt
dt
+.JW1 1.Zs -
(10.103)
correspondingly,
(10.104)
and inserting Eqs. (10.103), (10.104) into Eq. (10.101), the stator voltage equation in the moving reference frame results,
(10.105)
~ (L R 1iR + Lo 1is) + j (W1 -
w) (L R 1iR
+ Lo 1is)
mM (t) = ~ Lo 1m [is (iR e j ê)*] = ~ Lo 1m [lis ei R )*]
Lo mM (t) = -2 1m 3 (72 L s L R
mM(t) = =
These equations may be simplified by substituting the flux vectors, defined in Eqs. (10.25), (10.32) which are also transformed into moving coordinates, ./.
-j À
j 'f..R = ./. !:f.R e
1./.
= Ls I 'is + LI' o iR, = LR li-R
(ê-À)
(10.107)
+ L o_s li .
(10.108)
Hence the current vectors are
1·z s_- -Ls 1- (1./. -s 1· _ L1 (1./. -
'I-'
(7
ZR- - -
-
(7
'I-'
-R
R
1./.) + (7R -R 1 1./. ) 1+
- -1-
1
---
'I-'
(10.109)
'I-'
(10.110)
(7s -s
Inserting Eqs. (10.107-10.110) into Eqs. (10.105, 10.106) results in d 1./.
T' s -d-1:...s t
T/~ W 1",\" ,
+ [1 + J. W1 T S'J
111.4) lU-H -I-
rI
I· j
() 'l-'s - - 1- o 1 4) I = 1" l+(7R - I. . .<; Us t ,
1./.
(WI - w) TRJ ''1/'/1
(10.114)
(f ; ,'
-s
1 ) (1 -R 7jJ - -1- 1 7jJ ) J . 1+
- -1- 7jJ 1 + (7 R -R
*
(7 S
-s
(10.111)
~
Lo
3 (72 Ls L R
~ 13
(7
(7
1m [17jJ 17jJ*
~ 1m Lo
-R
-S
+
(1
1
17jJ 17jJ* J
+ (7s)(l + (7R) -R
(10.116)
[17jJ 17jJ*] .
-s
-S
-R
This is inserted into Eqs. (10.52, 10.53) for the mechanical motion. To arrive at dimensionless real equations, the transformed flux vectors are nOw normalised and splít in real and imaginary parts, representing the components that are parallel (x) and orthogonal (y) to the voltage vector. The flux linkages in steady state operation with sinusoidal voltages and nominal frequency, as indicated on the nameplate, and neglected stator resistance serve as flux references, 3 .12 Uso [xs(t) 17jJ -s (t) = -2- Wo
+ j ys(t)] ,
(10.117)
3 .12 17jJ 2 Uso Wo [XR(t) -R (t) =
+ j YR(t)] .
(10.111'\)
Tlle magnitude of the voltage vector in Eq. (10.98) is also referred to Illagnitude at nominal operating conditions,
3.12 Us(t) 'Ils(t) = -2- Uso Uso .
I,IH '
(10.] I!))
II. is llutccl that no restrictions with regard to waveform or frequency of t.hc
J
I .1
[(1 7jJ
(10.115)
= O.
(10.106)
'f..s = 'f..s e
(10.113)
are transient time constants. The expression for the electrical torque, Eq. 10.48, can be written in the form
The rotor equation Eq.(10.51) is transformed accordingly, with Y.R(t) == O,
1./.
LR
Since only the mixed terms of the product contribute to the imaginary part, this reduces to
d
+ dt (Ls 1is + Lo 1iR ) + i W1 (Ls 1is + Lo 1iR ) = us(t) .
RR 1i R +
(7
the torque is independent of the coordinate system because the inverse trans formations with e jÀ are cancelling. Hence, using Eqs. (10.109, 10.110) we have
. 'ê) d 1i . 1. d(lR e) -dte -jÀ = ---=!}, dt + J W1 iR,
Rs 1is
Ls
(7
T~ = Rs ' T~= RR
l d(. _j>..) di s _ jÀ .d>'. -jÀ d is --=-zse =-e -J-zse dt dt dt dt -
193
'1/' "
0.,
(10.112)
voH'lgcs lmve been imposed so far. WiI.h 1.11('1:(' ddillitiollS thr V()Jl.al~(' (~q1\atiolls (10.111, 10.112) and thc e.' ·
:i/,;II,()I'
pr":ir:i"ll
rOl'
LIli'
I,")'qll('
1-;'1' (IO.llli)
: 1.':: :11111<'
1.11(~ r()llowillf~ fonn
I !11
10. Symmetrical Three-Phase AC Machines
1'8 ~ (xs + jys)+ [1 + jWl T;J
) = Wo T -U Us + J'YR s 50 ' I
+ (TR
~ (XR + jYR)+ [1 + j
(Wl - w)TRJ (XR 1
= 2 (1 + (Ts)mpo (Ys XR -
eo
(10.120)
, dxs T s - - = -Xs dt
+ jYR)
.
+ - - - YR, 1 + (TR
XSYR) ,
(10.122)
, dXR 1 T R - - = ---Xs - XR dt 1 + (T 5
1 W2
SpR Wo
Pull-out slip at nominal frequency and Rs = O
RR 1 SpR = Wo (T L R = Wo T R '
(10.123)
(10.124)
T
_ Jwo mpo
(10.125)
m -
instantaneous normalised frequencies
"t.at.or frequency
Wl Wo lIlecltanical rotational r['(~qll eIlcy (speed) W
Wo I'U lor
frcq U(~llcy (,);\ (,)0
WI
W (.III!
(IO.1 2Ci)
1
(10.127)
(10.128)
(10.129)
, dYR 1 1 W2 T R - - = ---Ys - - - -XR -YR, dt 1 + (TS SpR Wo
(10.130)
d
C~o) = 2 (1 + (Ts)(Ys XR .
eo d (eco) Rs 1 Sps = Wo(T Ls = woT; ,
1 Us(t)
+ -- -YR,
Tm
M(~chanical time constant
:1.11<1 lhe
1
+ ---XR + - - 1 + (TR Sps Uso
, dys 1 Wl Ts = - - - Xs - Ys dt Sps Wo
wltcre mpo is the nominal peak torque for Rs = O, as defined in Eq. (10.86). With the following abbreviations for
Pull-out slip of rotor-fed motor, nominal frequency and RR = O
1 Wl -Ys Sps Wo
+-
(10.121)
+ (TS
195
six real differential equations emerge for the direct and quadrature flux com ponents as well as motor speed and positionj is an arbitrary angle of reference.
+ J YS) = O ,
- -1-- (xs
mM
+ jys)
(xs
1 ( XR - -1--
TR
10.3 Induction Motor with Impressed Voltages
W
mL
Xs YR) - (w, c, t) , mpo
(10.131)
(10.132)
Wo dt Wo These equations may be represented in graphical form by the block dia gram in Fig. 10.16, where the integrators with feedback are redrawn to form unity gain lag elements. Due to the rotational symmetry of the machine the angles e(t), À(t) do not enter the right hand side of the model equations Eqs. (10.127-10.131) with the exception of load torque mL that may depend on the mechanical angle. The block diagram describes the dynamic behaviour of the symmetrical motor, fed by impressed stator volt ages independent of waveformj the mag nitude us(t) of the volt age vector is given by Eq. (10.119), its instantaneous angular frequency wdt) by Eq. (10.96). As special case with Us =const., Wl = dÀ/dt = const., the transient condition of the motor operating on the symmetrical three-phase supply results, where U5 is the RMS value of the line- to- neutral volt ages. Because of the transformatíon into the voltage based coordinate system moving with stator frequency, the steady state values of the flux components and speed, assuming constant load torque, are constant. The steady state values are derived by setting the derivatives in Eqs. (10.127-10.131) equal to zero; the resulting system of algebraic equations is equivalent to the results obtained in Sect. 10.2. Of course, under dynamic conditíons the Hux components are no longer CO!l:;tant beeause there may be transient flux waves which are linked to the Hlator 01' rotor windings and, hence, move with the angular velo city Wl or L,)~ «.I I é.) rdat.ive to lhe c()ordjnat.(~ s'ys t.
111I
II). :~ yllllll( , Lrical Three-Phase AC Machines
l1., {I)
10.3 lnduction Motor with Impressed Voltages
OJdt)
I
.!:!.§...
USO
197
mL mpo
Uso
Oh
úJ
úJo
úJo
XSYR
I--=,
mpo
.!!!.b...
Fig. 10.17. Simplified block diagram of symmetrical induction motor
mpO
the fiux components, for instance in the statori by eliminating Ys we find mM mpo
, 2
Ts
d
Xs
'
1
dxs
Wl
2 + 2 T s dt + [ ( Wo ) 2] di2 1 + S;s
xS
=O,
(10.133)
with the eigenvalues I+-t---ó+ - - - - I
YSXR
81,2
.!:!!.... OJo
OJ2
Wõ "' i ,~. I O. LO . lllock diagram of symmetrical induction motor
r.."
h.v illq .r<'.'i.'ied volt ages of arbitrary wave form
.J1CIII.S OL'Lppropriate frequencies; as a result there are torque and speed iill.ioll S.
= Wo ( -Sps ± J. Wl) Wo
(10.134)
hence the pertinent transient is a damped oscillation with the angular fre quency Wl, belonging to a fiux wave that is linked to the stator winding and thus remains fixed in space. The damping is achieved by the dynamic short circuit through the volt age source at the stator terminals. Likewise the lower feedback circuit in Fig. 10.16 is found to be described by a differential equation with the eigenvalues
t. 111111< Vi l.!
WII( 'JI ill.'i pccting the general structure of Fig. 10.16 it is recognised that iI. (': UI I>(! I"(!
illvolved , which is due to the fact that the currents in the rotor winding are "I:lglld.ica lly induced rather than impressed through a galvanic contact; also, 1.11(' all[.';Jc between the stator- and rotor-ampereturn waves depends on the op(!lal.illg condition, whereas it is essentially fixed by the field- and armature i i , ( ~S iII th e case of a compensated DC machine. The presence of AC quantities ;t./ :;o 1.( ~11(1s to blur the view, even though this has been alleviated by the H'( I illa Le transformation. W h iII! 110
83,4
=
Wo
(
-SpR
± J. W2) Wo
(10.135)
The associated physical effect is a flux wave linked to the rotor winding which moves with slip frequency relative to the coordinate axis. Considering now the special case Wl = W2 = O, when the mechanically locked motor is fed with direct voltage, and taking the dashed cross-couplings iI/to account, two new non-interacting feedback loops are formed, where each (~lHbraces stator and rotor. The differential equations for transients in the :,:-axis may be combined to ,
,d?-xs
TsTR
-2-
dt
'
,dxs
+ (Ts +TR ) -
dt
+axs = O .
(10.136)
(·~igenvalues are negative real, belonging to aperiodic transientsi a cor IT:'; POlldiug equation is valid for the quadrature axis of the machine. Clearly I.II1'S(: ('qllal.iolls eorrespond to those of two-winding transformers, describing I.I/\' d l'(,()l/l'bll" ilW \( ~ l.ic illl.(~rad: ions in the two axes ofthe machinei under the ,dnl/lIlI't! 1,('1:1, ('llIlIlil.i"IIH ;1.11.1 ill vil:w or I.lw voIt.a.ge-based coordinate system,
!10th
19i;
10. Symmetrical Three-Phase AC Machines
10.3 Induction Motor with Impressed Voltages
5 p = 0.10,
'm (!o)
(01
(00
~
jO,5
(02 '=
'
(00
0,05 '=
const
0,15
-0,1
Re (!o)
Fig. 10.18. Root locus of induction machine for at different stator frequencies and speeds
W2
= 0.08,
= 0.05 =
the :/lux in the quadrature axis must be zero, Ys = YR = O, hence there is no torque. This is in agreement with the practical experience of a symmetrical induction motor with DC excitation at standstill. When analysing the block diagrarn in Fig. 10.16 more closely, i.e. for Wl, W2 f:. O, the mutual interactions of the two damped oscillators, coupled dircctly and through the speed, must be taken into account. This leads to a complicated resonant system of 6th order, the response of which is diflicult 1.0 assess. However, assuming Wl =const. and W2 ~ const., i.e. large inertia, l.hcre are considerable simplifications because the multiplications with Wl, W2
= 5 pR
= 5p ,
(J"S
= (J"R = (J"o,
T;
= T~ = To ,
(10.137)
tbe characteristic equation pertaining to Eqs. (10.127-10.130) is
[(To H 1)' - (1- u) + ;; (:;)']
+
1-
(J"
52 p
(Wl
+ W2)2 W
2
o
[(To H 1)' - (1- u) + ;; (::)']
_ O -.
(10.138) Tlle
Wo
= 0.05
= consto
(10.139)
consto
W2
5 ps
W2
The result is plotted in Fig. 10.18 in the form of a root locus in the complex s - plane, omitting the conjugate branch of the locus. The shape of the curve indicates that there is little interaction between the two oscillators in Fig. 10.16 for Wl » W2, approximately leading to the "decoupled" eigenvalues mentioned before; on the other hand, the interactions are strong at low val ues of the stator frequency, Wl ~ W2, resulting in slowly decaying transients. Figure 10.19 depicts the decay of a :/lux component at different stator fre quencies, assuming the sarne initial conditions; the transient torque is also shown. There is a remarkable effect of stator frequency and speed on the dynamic behaviour of the nonlinear planto Therefore the induction machine, when operated in a wide speed range, represents a very difficult control plant, which calls for special control procedures; some will be discussed in Chapo 12.
j1
1
(J"
199
solnl.iOI\S or I.hi~ (~q1l'ati()n, i.e. th~~ ('igcllvalllf's or (.h(~ systf:tn, havc heen
cOlllplll. ed 1l1l111 \'rir:dly 1(lr di/fcJ"(~nl. valtll's or :JI.III.()r rn~qll( ~ l1C:y WI, a.:;S lllllill~ !.II( ' fiol[\l wl l'IJ'. 1,III'hu lC'!..'r:·:
tlxs
Wo
= 0,05 = const tlmM
-
w1
m~i Fig. 10.19. Flux and torque transients for at different stator frequencies and speeds
W2
= 0.05
= consto
As a qualitative test for the validity of the mathematical model of the induction machine, which was derived with considerable simplifications, a line start at nominal volt age and frequency has been computed by numerical intcgration of Eqs. (10.127-10.131); the load torque is initially zero but, aftcr stcady state speed has been reached, half the peak torque is applied as load. Tlte results are plotted in Fig. 10.20, assuming a low value of inertia so that I.!te start is completed in a few periods of line voltage; the other parameter:; are the sarne as before. Besides the flux components in direct and quadrature direction, the stator Cllrrents are of main interest. The associated vector is produced by inverne 1.1':tIlSrolllliltion with the help of Eq. (10.109)
200
10. Symmetrical Three-Phase AC Machines
10.4 Induction Motor with Unsymmetrical Line Voltages in Steady State
201
xs{t} I I I I I I
~ H ~.
II
t' 11111i •
"'o
a
IIIIIIIIII'~(\.~----------t-------
'Idlll" ~ II I o ,
w
~ "'o
I I
!~~'Ys(t) 151
W
"I
I
I
I
I
I
I
I
Wo
I; \ "
/! /)
\ mM
1m \ po ~
(.
r·\ ../·.. . ·------
,/
"
.!!L(t) /
'
n
r./ úJo .-.-1. ,_,_._,_ './ ..... - . 1' mM
~.-
I
I I
mpo
Fig. 10.21. (a) Simplified block diagram of induction machine, (b) calculated transients
----
/'-
mL mpo
I
Sta/1 at no load
b
Load forque applíed
1
m L ="2 mpo Fi,,, . 10.20. Simulated starting transient of induction motOr wil.lt I
= li-s e j wd =
iv (t,)
,.
:~
12
~ .- -
2
Uso Wo (7 Ls --
[(
_1_ [1'IjJ (t) _ _1_ 1'IjJ ] (7 L s -s 1 + (7 R - R xs -
_1 1 + (7R
XR)
+j
ejw1t
(YS _1 +_ 1 YR)]
ejw1 t .
(7R
(10.140) Wil,h
r~ q.
'i:>i (l)
(10.5 5) follows
= 1Re [is(t)]
J :~ (/ '~'II (Ai l ) ri
/, ::
f(
:/:,'; -
1 - Xli ) CO:-) W I/' -1 I {r ll
eIS
1 (T
li
lI n
) Sill Ü)lt'l, c
(10 , 111)
This is also plotted in Fig. 10.20; it shows the characteristic large inrush current which, as the motor approaches synchronism, settles down to the small sinusoidal no-load current until the load torque is applied, The currents in the other stator phases are similar. There are strong alternating components of torque which also affect the speed, even though with reduced amplitude because of the mechanical inertia. If the inertia is predominantly on the load side, the shaft and the mechanical coupling may be severely stressed during the start. The speed, as a function of time, follows a similar curve as obtained in Sect. 3,2 when the electrical transients were totally ignored, Fig. 3.12; this explains also the absence of the overshoot in speed which is present in Fig. 10.20, As an intermediate step between the quasi-steady-state model and the more complete model of the induction motor, the heuristic block diagram in Fig. 10.21 a may be helpful. It differs from the purely mechanical model by the inclusion of a lag T~, following the steady-state torque-slip-function. For small values of slip, where the torque function is linear, this clearly re sembles the block diagram of a De machine. A comparison of the starting transient computed with this model shows good qualitative agreement with Lhe transiellts obtained by integrating the Eqs. (10.127-10.131). Of course, Lh e os(:illa(,iom; Hl1perimposed on speed and torque cannot be represented but I.I)( ~ lJl( :all viI,I'WH ! ~ i'( : wdl luo
202
10. Symmetrical Three-Phase AC Machines
10.4 Induction Motor with Unsymmetrical Line Voltages
10.4 Induction Motor with Unsymmetrical Line
Voltages in Steady State
U5 + --
3"1 [U -51
+ eh -52 U + e_53 -h U ] ,
10.4.1 Symmetrical Components
U5 - --
3"1 [u -51
+ e -h -52 U + e_53 h U ] ,
For intermittent duty drives, such as cranes or hoists, where simplicity of hardware is more important than high efficiency, induction motors with un symmetrical voltage supply may be used. By way of example, a phasor di agram of sinusoidal but unsymmetricaJ line-to-neutral volt ages is shown in Fig. 10.22 a. Due to the isolated neutral of the symmetrical stator winding, the sum of the instantaneous phase voltages is still zero, Eq. (10.37); hence the phasor diagram forms a elosed triangle,
fl.o
e -fYl}se -jy l}s+
b
fl5-
u5 ++
U 52 = e-h U 5 +
U 5-
!)S
c
+ fl.o ,
211" + eh U5 _ + fl.o , 1=3'
=t
U 12
[U 12
(10.14G)
eh U 23 ]
-
+ U 23 + U 31 = O
u 23 = (10.143)
ehfl5++e-hU5_+fl.o,
where U 5+ is the positive-, U5- the negative- and fl.o the zero-sequence volt age component per phase. The symmetrical components fl5+' U 5 .- form two symmetrical three-phase systems oE oppositc phase-sequencc. Solvilll( Eq. (10.14.1) yi(dds
(10.147)
is valid by definition. General1y the use of line-to-line volt ages does not ex elude the appearance of an arbitrary zero sequence component fl.o but, as seen before, it vanishes with a symmetrical load. If the motor is fed from a line possessing noticeable internal impedance, the symmetrical components of the terminal volt ages depend also on the load, i.e. the operating state of the motor. Should the voltages U 12' U 23' ~1 form a symmetrical three-phase system, e-j,,!
U 12
(10.148)
,
the result is
-
U53 =
(10.145)
Two of the line-to-line volt ages are sufficient to determine the positive ,Uld negative sequence components, because
It is known, that an unsymmetrical three-phase system may be de com posed into "symmetrical components" defined by
=
+ e-h)]
and correspondingly
Fig. 10.22. Decomposition of unsymmetrical balanced three-phase system
into symmetrical components
U 51
v
=0
-- 3"1 [U -12 - e -hu] -23
~s+À
~S2
!L23
!L12
The sarne holds naturally if the stator windings are Ll-connected.
!)S1
O.
1[U' -51- v- -U- 52 -e-h (fl52 - u 53) + u 52(1 + eh -' '--v---'
(10.142)
ejY!)s+
(10.144)
The expression for the positive sequence component is similar to that for the time dependent volt age vector, Eq. (10.34). Since the volt ages between the motor terminals and the isolated neutral are not directly available (the motor may be Ll-connected), it is appropriate to introduce the line-to-line voltages instead. By expanding Eq. (10.144) a we find, analogous to Eq. (10,38),
U 5+ =
fl 51 + U52 + U53 = O .
a
1 [U 51 + U 52 + u53] =
=
203
U -5+ -
1 e- j7r / 6 -12 U J3
-U - -51'
U-O - 5- -
.
(10.H!))
'l'hc case of a symmetricaJ induction motor fed by unsymmetrical snpply vol (.agci-i is fully containecl in the thcory developed in Scct. 10.1 bllt til<: sllb-· ~Cq'I('Ilt dis(".I\~;:;i()Jl oE som(~ cX
204
10.4 Induction Motor with Unsymmetrical Line Voltages
10. Symmetrical Three-Phase AC Machines
u (t) = 3 v2 [U U' e-i Wl tj -s 2 -s+ ei Wl t + -5-
(10.150)
,
a result, that can be interpreted as the superposition of two constant vectors rotating with equal angular velocity in opposite directioni hence the vectorial sum 1!cs (t) follows a periodic elliptical path, as seen in Fig. 10.23.
3..J2 U
2-
-s+
e jW1 t
components of Eq. (10.150). As long as iron saturation is ignored and the speed may be assumed constant due to a sufficiently large inertia, there is no interaction between the two symmetrical three-phase systems so that linear relations exist between volt age and current components,
Is+
= ys+ U s+,
Is-
3{2 -jw1 t 2- Us_e
w
= S and
Wl +W
ldS(t)
1
J
---
Y
(10.151)
í (t)e jo = 3v2[I e- jw1t j , 2 -R+ eiw1t+I* -R-
(10.152)
-lI
while the rotor current vector in rotor coordinates is, with de:/dt COllst. ,
+ 1* e-j (wd- w ) tj -R-
=W
"(J I./tl. i 111 ': wi!."
I'c'I '(II, 'lId l lll \ 1111 K d('II: d !.y WIL VI'1i iII
1+
1
j
Wl
Ls
.1
2-S
JuS;: 1+j 2 S
S
= 3 Lo 1m [15+ DH + Is- IR-l = mM+ + mM-
mM
mM+
mM
mpú
mpo
mpü
-=--+-
5p
(10.153)
I l
(10.156)
.
These two components of the mean torque may be derived by the sarne reasoning as shown in Sect. 10.2, because each is produced by a symmetrical three-phase voltage system. ln accordance with Eq. (10.85) we find
~ .
l+jl~ 17 sp . s Ls 1 + Jsp
2
'I'hc spced w i::; counted in the direction of th<: positive sequClIce field. Rence 1.1)(', frcqllcllcy 01' tlte positive and negativ(' ;' ('
(10.155)
When substituting Eqs. (10.151, 10.152) into the torque equation (10.48), four terms arise, two of which are alternating torque components having the frequency 2 Wl i contributions to the mean torque are only produced by interactions between stator and rotor currents of the sarne sequencei hence the mean torque, counted in positive sequence direction, is mM
w"cre Is t, Is- are the symmetrical components of the stator currents which ;1.1"<: ddined in accordance with Eq. (10.143). Likewise the rotor current vector iII sLal.or coordinates may be written as
t
i
p
ln steady state, assuming constant speed, similar definitions hold for the vector of the stator currents,
v2 [I eiw1t+I* e- iw1t j i.(t)=3 2 -s+ -s, -s
Wl
= S-
Fig. 10.23. Volt age vector of unsymmetrical balanced supply
(WI- W )
=2_ S
the admittances, derived from Eq. (10.72) with Rs = O, are
Y s+ = .
,.Ln (t) -- 3 v2 [I j 2 -R+ e
(10.154)
i
Wl
Wl
/"" -
= Ys- U5-
where Ys+, Ys- are the positive and negative sequence stator admittances per phase. Because of the effective values of slip for positive and negative sequence components, respectively, Wl -
///
205
+ spS
Us+ Wo [
U50
Wl ]
2
2 2-S
5p
+~ 2-S
Us- wo] 2 [ Uso Wl
,
(10.157)
where mpo is again the peak torque, Eq. (10.86), at nominal voltage and Sp the pull-out slip for Rs = O, Eq. (10.71). Eqnation (10.157) may be interpreted as the net torque produced by two identical mol.ors coupled to the same shaft whidl are supplied by syrnrnetrical tltn'('-plt":;I' voll.;!.J'," sysl.('llls I, US (lf (~Il'lid fl't ~ljlICm(:y hllt oppo:-;ite pltasc :;('
"s
\, iI'. I II '1\
206
10. Symmetrical Three-Phase AC Machines
ls+
!
1s-
10.4 Induction Motor with Unsymmetrical Line Volt ages
!
-h) _. 1 _ - 1( h U U -s+ - -s 3 e - e U 23 - J v'3 U 23 - U AM
w
Us+!
L-+----J
I )
IUS-!
,
207
(10.159)
where U AM is the line-to-neutral volt age of the disconnected line-phase. When combining the two equations, we find L-+----J
U -s+ - y
YsU Zs+ U + y s- -AM - Z s+ + Z s -AM, s+ (10.160)
Us_ =
Ys+ Ys+ + Ys- U AM =
Fig. 10.24. I nduction motor with unsymmetrical supply volt ages, equivalent scheme
Eq. (10.157) clearly demonstrates the basic disadvantage of alI control schemes employing unsymmetrical supply; while the torque is reduced, the power losses are increased by the negative sequence voltage and current com ponents. By changing the rotor resistance (Sp) and with different circuit connec Lions, resulting in speed-dependent positive and negative sequence voltages, a variety of torque/speed-characteristics can be obtained. This is discussed with the help of three examples. 10.4.2 Single-phase Induction Motor
zsZs+
+ Zs-
U AM
;
Zs = l/Ys is the impedance of a stator phase in symmetrical three-phase operation. The sarne volt age magnitudes would result if the stator windings of the two coupled motors in Fig. 10.24 were connected in series to the symmet rical three- phase volt ages. According to Fig. 10.26 the positive and negative sequence admittances can be taken from the normalised circular diagram (Fig. 10.8), where Is/Iso = Ys/Yso was plotted for Rs = O as a function of slip. This indicates that a large difference exists between Ys+ = Ys(S) and Ys- = Ys(2 - S) for small values ofpull-out slip, i.e. for a motor with a cage rotor. jlm(~) Yo
AI
Ys(S)
1 1 1
!..S1 =0
----L --,-
1 1 1 1 ;'
Ys +
y;;-=y;
------1
1.
a
R
Ys .(-) Yo
Starting capacitor
~
Fig. 10.26. Circular diagram showing positive and negative sequence admittance
;'
:l.C 'I Ih3= I}BC Fi!!:. 10.25. Symmetrical induction motor with single phase supply
Ily discolluec:ting one stator terminal of a Rylllluctrical induction motor, siuglc phase circ:uit in Fig. 10.25 rct;ul ts, A, II, n an' t:he terminaIs of the H.Y1l111I('!.ric:\.! lill(~ voltages. The cowll.,rnint, iUlpO!-lcd !Jy t.h(~ opcn circlIit is 1,11(,
1 "; 1
) '~"l l r :-: II V" j
lJ,,;
=H;
rUlld,II "! <'(111<1 11, ;<111 \; dl<1 1N11 rrnfll 1·:'1 J1 . ( 1.11 . 1<[1 ;) , ( IU. l·ll'i),
(10.1 ;)H)
If the motor operates near the positive synchronous speed (S ~ O), IYs+1 « IYs-1 holds, resulting in U s+ ~ U AM, U s- ~ O. This means that the negative sequence field is almost completely suppressed by the rotor wind ing due to the large relative speed 2 - S ~ 2; hence the motor operates with llearly symmetrical volt ages despite the fact that one terminal is not con nected. ln the vicinity of the negative synchronous speed the conditions are illverted. Th('! st.('ady-st.a.te torque of the single phase motor is obtained by substi LilLillg Eqs. (10. HiO) illlo Eq. (10.157)
208
10.4 Induction Motor with Unsymmetrical Line Voltages
10. Symmetrical Three-Phase AC Machines
I I Ys. Ys++Ys .
turned manually in either direction, the corresponding sequence prevails and the motor continues to accelerate at light load up to the operating speed. To make the motor self-starting, either some asymmetry has to be built into it, as is done with shaded-pole motors, where a short circuited turn is built into the stator winding, or a starting capacitor must be connected to the third or an auxiliary winding, as is shown in Fig. 10.25; in both cases the direction of rotation is fixed. The value of the starting torque depends on the design of the auxiliary winding and the capacitor. Besides the starting problems, the drawbacks of single-phase operation lie in the residual pulsating torque as well as reduced utilisation and efficiency of the motor which are always below that of a symmetrically fed motor. It is for these reasons that single-phase induction motors are only used at low power ratings, for example in household appliances.
I
I
Ys+ Ys ++ Ys.
2· Sp 2 S
Sp
Fig. 10.27. Symmetrical components of single-phase motor 11
10.4.3 Single-phase Electric Brake for AC Crane-Drives
.Q!.
úJ,
I
I
" mM+
I
, mpO
mM
I
mM· I
I
I
mpO
mpo I
I
I
I
_
m
I
I
----,---+
' ..... .....
209
mpO
Q
I
I
-1
3
Fig. 10.28. Torquejspeed characteristic of single-phase motor !l23 =
mM mpo
Fig. 10.29. Single phase supply of AC motor for braking duty
(10.161)
=
2
2
YS +Ys YS+ [ ~+s 1 sp --s
2
1
2-S s;-
+~ 2-S
IYs+Y:+yJ]
UAM Wo [
Uso
A simpIe intermittent-duty AC crane-drive required to operate as elec t.ri ... brake in the lowering direction may have the asymmetrical circuit COllllec~ tion shown in Fig. 10.29, where two stator terminals of a symmetrical wOIIJld rotor motor with large rotor resistor are short-circuited while single-phaH(' voltage is applied between the short-circuit connection and the remaillillf~ sta tor terminal [G23]. This is one of several possible connections having s ilJli lar properties. The analysis of the circuit is easy with the help of symmetricil.1
1
2
wd .
Since the torque rises with the square of the voltage, the negative sequence torque is even further reduced than the voltage. Some of the f1.Jnctions con tained in Eq. (10.161) are sketched in Fig. 10.27; by combining these curves the torque/speed characteristic of the singlcvlt;1.se motor is obtained, drllwll in the usual oricntation, Fig. 10.28. 1'11(' ClIJ V(~ is s'yl1llllctrica.l with rm;ped to the orif~ill hc("all s(: aI. S = 1., ys, = yC ; holdli alld I,\l(~ o[l[losiJl[!; t.orqU(· ("()[lIpOIl(·IIf. ~ 1
('11 1111<>1
O
i.I, ) 1I
U lInponcnts. J.llt,rodll<'ill{~ t.1l!:
('1 1.11<", ,1. 1I.'lIcl; lhe IlItlI.Ol" t!. ·V••I" p H IjO .1 ... 1, I.ol"tliw ;ü !; I.;lIId'!il.iU a li.! I, w l t.lulllf, (' xI,"l'lI n l ll Nl ii tl l,IIIU'O. Ir 1\ /1 11 11 111 ItIllt.OI :d, 1;l.:l.Ild nt.i\ll 1M
(/ I ::
/I
.·111'
cOllstraints ()f t.hc circlIle /I '.1 :1
.- {)
illl,.. I'~q t! (III H !,) , (10 I Iii) [, " :11111.: 1 iII • _.#'
(I().I(;~)
10. Symmetrical Three-Phase AC Machines
4 10
1
_
= Us- = "3 UAB
Us+
-
1
-v'3 UAM ,
10.4 Induction Motor with Unsymmetrical Line Voltages
(10.163)
illdicating that the symmetrical voltage components are of equal magnitude, illdependent of speed. Hence, the two machines in the equivalent circuit, Fig. 10.24, would have to be supplied by constant and equal three-phase voltages with two terminaIs of one motor interchangedj this renders the superposition (,f the two opposing torques particularly simple.
úJ
,,
, úJo 1
" I
,
mM+ ',m po
",
\
m M. \ fn po \
\
\
3
\
------"-\---
\
\ 1.
\ \ \
,, ,
3
,,
Finally, Fig. 10.30 also contains the torquejspeed curve with symmetrical voltages but inverted sequence and short-circuited rotor resistance (c). This operation allows regeneration, however at higher than synchronous speed in lowering direction. It could be used as the "Full speed, Lowering" set-point of a contactor-controlled AC crane drive.
10.4.4 Unsymmetrical Starting Circuit for Induction Motor Smooth acceleration during start-up is important in some applications, such as textile machines, where there is a danger of breaking fibre threadsj an asymmetrical circuit like the one shown in Fig. 10.31 can provide a simpIe solution for this problem. An external impedance ZI is inserted between a line phase and one of the stator terminaIs of a symmetrical cage- or wound-rotor induction motor; the circuit is one of several similar connections. Assuming the impedance ZI to be constant, all volt ages and currents remain sinusoidal in steady statej the following equations hold: ZI
\
\
_ 1.
(b) Braking with De exeitation
\
I Sl + U 12
m
mpo
211
U 23
= U AB = (1 -
= U BC = (e-h -
e-h) U AM ,
eh) U AM .
(10.164) (10.165)
(a) Single phase braking A
,,
-----~
-1
':'~--\;)S;m~etrieal voltages reversed sequenee
"'l IJ:. JO.30. Torquejspeed curves of AC motor in unsymmetrical connection
'l'he torque curves are depicted in Fig. 10.30 assuming Sp = 2.0 caused by ;\. large external rotor resistor. Due to the reduced voltages the peak torques i UT only of the nominal peak torque, hence the motor is again only partially IIl.iliscd. Since the resulting torquejspeed curves lie in quadrants 2 and 4 the drive can only operate as an electric brake, regeneration is not possible with I.I.is scheme. An advantage of this connection, when used as a lowering brake in crane applications, is the fact that the resulting torquejspeed curve (a) passes I.I. rough the origin; unintentional speed reversal (raising) at light load does 1101. occur as would be possible with symmetrical supply and the rotor resis I.allcc sLill furl'.hcr illcreased. SillJilar hrakill ~: dmfacleris(,ics aJ"(~ liI/I.il.ill,'d with H sllIaller ro(,or J"(~SiH1.1~1I1·" 11'y r,'('djllf~ I,II!' til,nl.ol" windiJlI{ wi l.It ,11,'("<'1, (' lI.rT '·.fJl. (h). TIl(' lIlol.or 1''''' 11
k
"1' ( ' 1' 111." ' 1 rl' i 1111
\'dd ,v
1' 111 ('(' 111.
I,r ll.h'.
!J23 = !Jae Fig. 10.31. Unsymmetrical starting circuit
The s tat.or current can be expressed with symmetrical components, Eq. ( I O. 1~H), ;;, I
' :i I
;;, I
1\',.,' I 1/ ,'.,' I I Y·; (l ·.' .I ;
(10 . I (iii)
~!
10. Symmetrical Three-Phase AC Machines
I ~~
10.4 Induction Motor with Unsymmetrical Line Volt ages
likcwise the line-to-line voltages obtained from Eqs. (10.143) are U 12 = U 51
-
U 52 = (1 - e-h) U 5+
+ (1
U U - -53 U -- (e -h - eh) -5+ -52
U --23
maps the real Rl-axis into a circular arc in the complex plane, extending from RI = O to RI -+ 00. When the rotor is running, O < 8 < 1, the phasor diagram changes with the tendency of reducing the volt age R1I s1 , thus achieving improved symmetry. This corresponds to the situation found with the single-phase motor.
- eh) U 5- ,
+ (eh - e -h) -5U .
811bstituting these equations into Eqs. (10.164, 10.165) and solving for U 5+' lT. s _ results in
3 + Zl Ys U -5+ - 3 + ZI (Ys+ + Ys-) -AM U
-
u
=
-5-
(10.167)
,
-Zl Ys+ U. 3 + Zl (Ys + + Ys-) -AM
+ Ys - us-l
= -3 u 5-
(10.169)
is a measure of the asymmetry of the stator voltages. With these results the phasor diagram of the motor can be drawn for any chosen operating point. Tlte torquejspeed curves may be computed from Eq. (10.157); their shape dq)ends on the impedance Zl and the motor parameters, but they lie inside i.Ill~ :IIea which is bounded by the curves for Zl = O and Zl -+ 00. This is qll:dil.atively indicated in Fig. 10.32 for a low-power induction motor having I'ni rly large pull-out slip and assuming a resistive impedance, Zl = RI. Of ('1111 rsp, this and all similar arrangements are associated with additionallosses iII t.hc external components as well as the motor itself and are only suitable ror short duty operation. The function of the circuit shown in Fig. 10.31 is further illustrated by I I iS!:\lssing the phasor diagram of the stator volt ages in locked rotor condition. AI, 8 = 1 we have Y s + = Ys- = Ys(l). Hence it follows from Eqs. (10.167 10.169) with Zl = RI =
R I .
.L -SI
3 RI Ys(l)
3 + 2 RI Ys(l)
as a limit.ing case .'
hm (l{IL"; I)
U1
wil.11
H'XI
1.11<'
(RI -+
O :S RI
< 00
(10.170)
;
this includes again the single phase motor
= 'l U AM , 3
(Ir thc f;l.al.or vO! l.nW·!4 ai, ,<.,', IO ,:I:\) 1':'1 , ( 10 , I'/(l) 1'1'/11' 1'111 '111,:: li,
1l11 1l.':oJ' d i a l'/H III
:. 11 1. 1' 11,11',111.1 111 1 '
00)
U -AM ,
( 10'11 '\ '
úJ úJo
(10.168)
'l'ltis contains the limiting cases of the symmetrically fed motor (Zl = O) aud the single phase induction motor (Zl -+ 00) discussed in the previous paragraph. ln general, the positive and negative sequence voltages are com plicated functions of the motor admittances which in turn depend on speed and pull-out slip. The volt age across the series impedance Zl I S1 = Zl [Ys+ U 5+
213
I dq(' 'II ('I' tll. i 111':
mM mpQ Fig. 10.32. Torquejspeed curves of unsymmetrical starting circuit
for large pull-out slip
A {\--Rj=O
ZI=R I
S=1
\AB AR
3,C
!J Be
R I =oo
2,8
Fig. 10.33. Phasor diagram of asymmetrical circuit at standstill
The shaded area AR of the triangle (Fig. 10.33), formed by the terminal voltages fl 12 , U 23' ~l at standstill (8 = 1), can be expressed as
AR = ~ 1m [fl23 U;ll ;
(10.171)
lilll' llI'
o
1111,(1
1'1I11.-1.i'lll wil i eh
whjeh hy
ii '"
(10.172)
illsI'l'l.illl'; Eqs. (10.145, 10.14ô) ma,y be written as I,;
III}
1
(!?, I, "
(10,'173)
;, I '1
I () .
SymmetricaI Three-Phase AC Machines
A comparison with Eq. (10.157) shows that AR is a measure of the torque at standstill [28J; it vanishes for Rl -t 00 and assumes its maximum value at /lI = O, i.e. with symmetrical supply. The circuit shown in Fig. 10.31 can be extended by additional controllable i III pedances to allow continuous speed control, even four-quadrant operation hlll. llot regeneration. This has been realised in the past with saturable reac lo!s, the impedance of which could be varied by direct current in specially c , >llllccted control windings. Because of the high losses these drives were re ~;I.!ictcd to short duty operation; also the control was rather slow due to the si 11J';gish response of the large reactors, making this type of equipment obsolete I.<>day [L29J. A similar control effect, however with greatly reduced size and ("o s I. of the components as well as much more rapid response can be achieved hy illscrting antiparallel thyristors into the stator supply leads of the motor; l.!t is causes considerable nonlinear distortion of the motor voltages, resulting iII yet increased losses but it may be acceptable in short duty applications.
11. Power Supplies for Adjustable Speed AC Drives
An important result of the preceding chapter is that the application of AC motors in continuous duty adjustable speed drives calls for static converters of adequate power, generating three-phase volt ages of variable amplitude and frequency. This is necessary to maintain at all speeds a low rotor frequency, which is a precondition for acceptable overall efficiency of the drive. Con verters of this type are available today employing thyristors, gate turn-off thyristors (GTO), or switched power transistors (MOSFET and IGBT), but the complexity and cost of the converter equipment and control still exceeds that of line-commutated converters of a similar rating. While part of the cost for the converter can be recovered by the reduced cost of the AC motor as compared to a DC machine of comparable rating, the overall cost of the AC drive may still be somewhat higher than that of a standard DC drive. Of course, AC drives exhibit a number of important advantages, most promi nent in the case of an induction motor with cage rotor, that will inftuence the decision in their favour: • Power rating and speed of an AC motor is not limited by a mechanical commutator, • Reduced axiallength, volume and inertia of an AC motor, as seen in Fig. 10.1, • Rugged design, reduced maintenance and service requirements, • Drive is applicable in explosive and contaminated environment. The last features are of particular interest with traction and servo drives where many driven axes are incorporated into the machine tool or robot. Of course, there are also applications at high power and speed, where DC drives could not be built and only AC drives can provide a solution. Since the cost proLlems will diminish as the converter- and AC drive- technology matures, it is likely that DC drives will eventually be completely superseded by the more advanced AC drives . The basic lay-out of an adjustable speed AC drive is shown in Fig. 11 .1; iI. collsists of an AC motor, a static converter, which generates a variable voU.age, variable frequency AC system and the associated control equipment. 'I 'II< : i 11(".)"( ':1:-:«:<1 col1Ipkx:it.y of the c:onverter is callsed by the fact that inver I."I'Hwil.h fOI'( ·, 'd ("() li 11 11 II Latiol1 may n:q\lÍn ' :ul
11. Power Supplies for Adjustable Speed AC Drives
}I(i
11. Power Supplies for Adjustable Speed AC Drives
217
'J'1I1,l c 11.1. Adjustable speed AC drives, synopsis
° W';3
e-
~ ç1 ;.::::
~-:.
~iSal~ 0"",0.0.
m
~
°
u ;,.,. ". 0
~~
~ro~
~~]~~~~~
0.2:.-::; .... 0.2: (fJ d ~>, ..<:1013].<::
0.,<:; '" • ..<:11.()~ç1 .~~ 0.8 :::r::~.....:l "d
8..<:1 13 2'~ 13
u
s....
S
~-ro~
1'1
UJ
tJ'"'
bOo;':::: Q) .bOl.() O"d
~2-
r;;.~~....:l"d
o o
o o
bO
ContraI
bO
Fig. 11.1. General scheme of AC motor control
(fJ
•
I
~ t) J; Q)
t-' .~
~
H
.... Q)
;>
13
'>
§~o
.... "" Q).3 ...c:: > o
Q)
-§,o~-5-§,gbO§S
~~$~~2-~9
. <:1
o.
;:l
~
'"
~
. <:1
UI,
.
°
.... J; "
O)
..
<.J
I e- ~§"'~ U
..
o;>~..oo"",o. ~::g(j o..~ VJ
..<:10 -+J
;:l
g 00 ~~] ~ ç' o.:: ~ ~
~
S
H
I
.., S
ç1::;~~S
~t;O~o
..... - t-< ....,
U
UI;...·~;>
U
::Jt::,CJ ~O)
'tt1" I
8
"
.... oç1 tE
o
,"
.
1- ,
~
,.)
u '
....
Q) ~
n' -,'
;~
:~
i).1
r
I"
. <:1
O)
lJl
2:
..
O)
~"d
~Ç'o
(I)
0;>0
g~ 13 m t) ;:l
S..,;.a .... 0Q)
O)
~ ~ 0...>'< bO.2 13 13 00 ~oC~oo 0.0 or-lQ)~;>~ .-<
~,... . .- -u° e-<
.....:l~;>
. ~ ~
~~o
"d
....
'"J,
S ~
'1'1
I:
°
I
>=l
d (l) ç1
'u
~
(,,1
VI
O)
R
0.:
° °
....,
ç1
0
ç1 O) V>
o r.l,+'
.~ 's-s ~ w.;,l
p'i. 0',
,. , Ir;
I~
-
~
~ .... ro
te
~ § t)
...,+-> u o
E S
~
f1
. 1-
S~ .... '"
'-'
Uo '
8.~ al ~·C
2: o. o."d 8 "'dO) 3.-<..<:1 >,~ ;.a V bO "d ç1 Q)
-:B ro
L -_ _ _
constant
Commutator / machine-side converter
I
DC - link
I
I
1 -----+-1 ...._ _ _ _---I.
~
bO
AC - motor •mM' W
> UI' úJ
0
1
úJ
1
I variable
I
.... o
Õ Õ dO) .... I
bO
I
ç1 m "d
o U
'..sª
"a
úJo
o~
.... o
>=l
Line-sde converter
° lJl
2:
/
~.--_.-.-._-----_.----
~~"--::,",,.2~
.. ~
DC - machine
Supply
..
O)
~''' ~. ~ §' ~....
V"
~§~
"d
(l)
~
:1
110 ""'" ~
:.a2:~ .2~~
~ .&~ ~ ~ 2:
"foiS .~
32<..<:10 .-< ......bO"d .'"d - '-"' . <:1
,íl, '--,
§
° ~oo_
0...--::.
o-----.:;..~
o.iS ô.-::; 8 ~ o. '"
• I
....
~
o
~ ~ ~
U . ~/ ~ . ~I oI-' ;:. . II, ', I'~' Ü~ UOr1 I " ,
O)
....
Q)
of the inner torque controlloop which is much more complex than for a DC motor. The reason is that an AC motor with its simple mechanical design represents an involved nonlinear multivariable control plant.The motor must be fed with alternating currents of variable amplitude, frequency and phase; in addition, the rotor currents cannot be measured with ordinary cage motors. This is in contrast to a DC machine, which has a very simple decoupled control structure but a more complex mechanical designo The mechanical part of the control scheme is, apart from the smaller inertia of AC motors, the sarne for DC- and AC-drives. Fig. 11.2 indicates that the mechanical commutator of a DC machine acts like an electromechanical converter which feeds the armature conduc tors with alternating current of a frequency proportional to speed, Fig. 5.2 . On an AC drive, the functions of current conversion and torque generation are mechanically separate, thus creating new options for electromechanical interfaces .
'.;:Z~
ç1 ;:j
U "' O
.~ ~ I~
"d
o....
.. >") ° .3
0)....,
4-<
H
-"d
o
~o § 8 O ~
o
-,
~-------------- ______ I
.... .2o ~v ç1bO d ç1 .
Fig. 11.2. Controlled AC-drive with DC-link converter
>=lm"d
~ .-d .§
0-
IS'
~ ..<:l
j;)"
§ ,?:l t
Table 11.1 presents an overview of controlled AC-drives, showing five ba fonns of power electronic converters for changing from the fixed line- to v;l.rial'](', 1I10Lor" frcqm:ucy. FOlll' of til(: circuits contain a DC-link for decou 1'1 i "I', 1.11(' I.W() :;id4'1 : 411' I.!t\: CO \lVerLCT , w 114'1'( ';\:; I, h( ~ cydo-collverter performs this
:;ic
I ],'~ .~ II )
--'
"114'1 ii. Li 4111 iII
; 1.
:1 ;111'..100 C'O\lVC'rnillll
:il,;lI'.4' f\ II 01' 1.114'
4'nllvc'rl.4'l':-1 are
;t.dapl, ;~hk to
11.1 Pulse width modulated Voltage Source Transistor Converter
11. Power Supplies for Adjustable Speed AC Drives
218
four-quadrant operation, they could also be used with variable speed genera t.ors, for instance in wind energy plants. ln the vertical column of Table 11.1 live basic types of AC-machines are listed that may be used for controlled drives; with the exception of the reluctance motor, which is characterised by salient poles, they can alI be treated with the mathematical model derived iii Sect. 10.1. ln principIe, most of the 25 possible converter-machine combinations could 1)(' realised but only those, where comments are entered, are presently applied i II practice; since each type exhibits particular advantages, it is unlikely that I.h(' lield will eventually narrow to just one or two major circuits as has been t . It(~ case with DC-drives. ln fact, new combinations may be added as new ~;lViLcbing devices and converter circuits appear on the scene. An example of Ih(' oIlgoing development are resonant-línk converters, presently restricted to Imv power applications such as switched mode power supplies; cyclo convert crs with forced commutation (matrix converters) are also a possible future opLioll. Only a few typical converter topologies that show promise to stay \ViII be discussed in this chapter.
Pulse width modulated (PWM) Voltage Source Transistor Converter (IGBT) 11. 1
TI\(~
rOIlf-quadrant DC/DC converter shown in Fig. 9.13 a can be extended by \,wo more arms, as seen in Fig. 11.3, to form a three-phase converter fo r 1."(' slIpply of AC-motors in the lower and medium power range, from ::11111.11 hiV;h d'yl1amic performance servo drives with speed and position control ( . I () I, W) Lo lIloSt auxiliary drives in industry, ranging up to several hundred li 'vV, II\. Itil;hcr power this two-Ievel circuit is also extended to multi-leveI ('oll/il',1I r:l.t.iOllS. The converter is suitable for supplying induction as well as Ily lI('h \( )1l01lS motors. W iLh Cite voltage source converter shown, the terminal voItages of the mo \'()f' ( '; \.11 assume the three possible values UD, O and -UD.When the motor is r('I'.( ·II('l' at.illg, the mean link current I D is reversed, calling for a two-quadrant :;lIl'pl'y of the DC link. As long as the power levei is low or when regenera Lioll OCCIll'S only occasionally, for instance during dynamic braking of a servo drive', Lhe DC link is normalIy supplied by diode rectifiers and the reverse I" ,wcr is clissipated in a balIast resistor, Fig. 9.14, as indicated also in Fig. I 1.:\ wiLh dashed lines. The pulse-width modulated ballast circuit then be I'.IIIS dischargc the link capacitor as soon as the link volt age rises above a pr('s('\. value. Of course, any other source of direct voltagc could also serve 1;)1 I'( '( 'dillg t.hp converter, such as a battery 011 an electrically driven vehick, wl\('re r(~I~: ('II( ~ rntion is il1lJ)ortant t.o sav(~ (m('rr~.Y. Wi\.h Il ipolnr t.mllsi st.!ll'K, Lhe :;wit.dlÍlI ~~ fr('qlwlI('.y i:-; lilllil.(~dto n r{ ~ W 1\1-/(" IlIll. \\'il.ll low 1"'111'('1' 11I:;III:t.I,(,·cI C a L{' 1111'01 111' ' 1'1' 1I. 1I11 íl~I . " rH (leIIT), H:-i :; ho wlI iII aclclillf~
L()
219
io
-', : : RB "
Uo
I
, I I' ,
.....
I~TB I I
iSI
iS2
iS3
iSI
Fig. 11.3. Switched IGBT- converter for three-phase AC motor (simplified circuit)
Fig. 11,3, or Field Effect Transistors (MOSFET), having a switching time in the order of a microsecond, the mean switching frequency may be above 16 kHz, i. e. beyond the audible threshold, so that no objectionable acoustic noise is emitted from the drive. An additional advantage is that the converter then exhibits a large bandwidth for controI. If the link volt age UD is of sufficient magnitude, fast current control loops can be designed which keep the stator currents in line with the alternating reference values; this effectively produces near ideal current sources for the stator windings of the motor, thus eliminating the effects or the stator volt age equation Eq. (10.50) on the dynamics of the drive. As a consequence, considerable simplification of the drive control is achieved because the stator voltage interactions are then dealt with by the current controlIers. These could either employ pulse-width modulators operating at constant frequency or simple On/Off comparators having a narrow hysteresis band, also indicated in Fig. 11.3. The resulting current waveform with a single-phase load and On/Off-control (symmetrical modulation) was shown in Fig. 9.16; it has the advantage of fast response but exhibits variable switching frequency which may be undesirable in view of interference problems. When extending the principIe to three-phase loads, as in Fig. 11.3, the hys t.eresis band is helpful to avoid interactions between the controUers because, in vicw of the isolated neutral of the machine winding, the stator currents are balanced and one of the controUers is redundant, it is only retained for sylllmetry. To produce a prcfcrred switching freqllency, a triangular clock signal may Iw added aI. LIt ~ ~\l\lJl\lillg poillt or t.Jj(' cmrclIL cnntrollers [829]. Another pos ::ihilit.y j:: 1.1)( , >1 YII('hlOllj'l,aLwJI (lI' UI(' (, ()IILIOII('\"~:, i.t:. :dl()wiJli~ a t,rallsj(.joll to
'2'20
11. Power Supplies for Adjustable Speed AC Drives iRef (I)
11.1 Pulse width modulated Volt age Source Transistor Converter
*j
-20~
'
7~;ef """---'
b
;f
No ,,,d
20
200 7
-20,
MM Load '
~;ef
'" F
)~,
I
I-
Stator current references i
A 200
221
, is
5ms
-I
~cc,n' ~
-200~
"
i à 200
F i(.!;. 11.4. Current waveform produced by a single-phase transistor converter with illdnclive load impedance and synchronised On-Off controller
only at equidistant sampling instants t = v To defined by a fixed clock Cn:qlwncy to = l/To [B45] . While Fig. 9.16 was derived with a simple free l'llllllillg On/Off-controller, the effect of synchronization is sketched in Fig. I 1,'1 , again for a single-phase converter with inductive load; the amplitude (lI' LIli: ripple superimposed on the load current is now increased, because the ::w; I.l'hillg op erations are delayed until the next sampling instants, thus cre :,",,;II/,; waiting intervals ranging between zero and To. If synchronised On/Off {'O lIl.r()III'I'~ a.re employed, the hysteresis band Ll can be omitted because there ,:' ""W :\. IlIillimum time interval To between subsequent switching operations. W hl 'll a very low ripple content of the currents is specified, as on machine I.,"I! CI",tI drives, the use of a pulse-width modulator in combination with i' 1""'; 1,1' I"llrrent controUer is preferable to On/Off current control. This is d l 'IIIIH, ~ I.], ; lt e d in Fig. 11.5 where measured current waveforms obtained with III<' :: ;1.111 " three-phase transistor converter are compared. Fig, 11.5 a depicts III<' ('IIIT"Jlt produced by a synchronised controUer having a sampling period 'I II '22'1,s, whereas the curves in Fig. 11.5 b show the result of linear current 1;0111,1',,1 with a pulse-width modulator operating at 5 kHz [H20].The audible 1IIIi :;I' is !llorc pronounced with PWM control unless a higher clock frequency i:: d()~I: JI, but the current waveform is much more acceptable. The reason for 1,111' illcn-:ased ripple current is the quantisation of the voltage-time area with 1.1)(' sYll chronised On/ Off - control. 1'lIbl:-width modulators are now available in a variety of integrated cir " II i!.:;, wltich gn'atly silllplifies the design of PWM COllverters, Thcre is also thc pnHs iliíliCy 01' so[l.wan:-based mo(lulatioll HHill/ '; fasl , si!!:J1 al proccssol'S which "II; ' I' III11illlil.('« 1I1'xihili I.y hy colllhilliJlI.r I'W M wi I.h H0l'hisLÍc,ÜI :d, I'udl as
-20
-200
()(TlIf
b) Currents with linear contrai, PWM at 5 kHz
Fig. 11.5. Comparison of current waveforms with different types of controllersi (a) Current waveform with synchronised On/Off-control, To = 22J.Ls (h) Current waveform with linear current controller and PWM ,
operating at 5 kHz
predictive or time-optimal current control; this is of interest with high power converters switching at a lower frequency [H67, H71], The assumption of virtual current sources for the stator windings is, of course, only valid as long as the ceiling volt age of the converter is not reached, This is demonstrated in Fig. 11.6 for a free running On/ Off - controUer when the frequen cy of the current reference is gradually increased. Due to the inductive load impedance, the necessary fundamental component of the volt age rises with frequency according to UI = h JR2 + (wIL)2, It is seen that, beginning with Fig, 11,6 b, there are signs of saturation as indicated by pro longed intervals where no switching occurs, which is due to insufficient supply voltage, Clearly, this results in a rising control error between the sinusoidal current reference and the piecewise exponential feedback signal and must be taken into account when designing a drive control system, A very effective method of pulse-width-modulation that is particularly slIitcu for fast switching converters is called vectorial PWM because it repre s(' IILs a n attempt to reproduce in a given time interval vTo < t :=:; (v + 1) To a v()l ta~t: v(~ctor 'U s IId (t) demanded by the current controUers [P29], For this iI. is illJp"rl.alll. 1,(1 rellle m!>!:r t.lt al. tll(: COll verter circnit of Fig_ 11.3, which in
·.! 22
;/
11. Power Supplies for Adjustable Speed AC Drives
.m.~; iReI (t), i(t)
/)
223
11.1 Pulse width modulated Voltage Source Transistor Converter
protective interval which lasts only a few rnicroseconds when transistors are used, can be assigned to a finite switching time in the converter model. When writing the volt age vector 1f.s(t) in terms of line- to-line voltages (10,38) which can assume the values UD, O and -UD , 1f.s(t)
" 2 = USI + US2 eJ"Y + US3 eJ "'( = U12
-
U23
-'
e
(11.1)
J"'(
u(t)
OOllPl~i
six distinct voltage vectors and two zero vectors result, as seen in Fig. 11.7 b, Clearly, when going from one comer of the hexagon to the next, only one leg of the converter needs to change its state,
1l.2
"Zero vectors' 2x
S1
Uo
S2
,,
S3
U
~ -SRef
'.
-1
(
\ .' U l -I
()ff
I I I I
I (.'
i(t)
u(t)
U23
"51 [
a
"'2"52 [
uS3
[
b
Available voltage vectors
Fig. 11.1. Switching model of converter in Fig. 11.3 and generated output voltage
, .,
"
vectors
,.,' iRe/t)
S;LI.llration effects of single-phase switched transistor converter with ( .,, / ( Hr cml't:1I1. control supplying a passive R-L load impedance 10' 1/\.
I I. H.
I",!" . I I. 7 a is represented by a switching model, must be so operated that 11t0l11' 0(' I.h(~ Icgs is short- circuiting the link voltage UD and that each motor 1.('J'llIill;d osition. Tlw switching :;t, at(~ (8 1, ,'!h, Sa) of the complete "4IIIVl'I'l.t'l' i ii !.Il(ili dcscribed hy a. tlm'c hit. biuary word h:wiug eight diffcrcnt V: dlWi l , ill(~ llIdilll': (I, I) I) aud ( I, - I, I) , (·"III,tI Y, I'I'O v( ~dor sl,at.(·s, wll(~rc t.11<' 1/101.01 1.1 'IIII ·ill ll.h 111'( ' li hol't. (~ir(,IIi1.('d "" . 1.111' 111'111'1 "I' I"wlil' 'ne IIml. Tht:
It is assumed that the converter is switched at a basic clock frequency = l/To, which may either be fixed or correspond to a multiple of the variable fundamental stator frequency, and that the current controllers have determined a command value 1f.SRef, valid in the next interval To; 1f.SRef is assumed to lie inside the first sector, as shown in Fig. 11. 7 b. Since this commanded voltage vector is normally not coinciding with one of the available,voltage vectors, it is to be composed by a switching sequence comprising the adjacent vectors 1f.l = 1f.S(Sl' 8 2 , 8 3 ) = 1f.s(l, -1, -1) = UD and 1f.2 = 1f.s(1, 1, - 1) = UD ej1T / 3 , while filling up the rest of the time interval with zero-vectors. The sub-intervals tI and t2 for the two adjacent vectors are to be computed from the following equivalence
lo
1f.s Ref
) h t2 U = (USa + J'USb Ref = 1f.l To + 1f.2 To = D
( tl To
+e
j
1T
/ 3 t2 ) To (11.2)
III
addil.ioll /, I I 1.1. 1- 2to
To holds, where to is the zero vector interval.
224
11. Power Supplies for Adjustable Speed AC Drives
t
Samplíng of current
!.- T. •
O
1
.. - - .-'
-
t 1 i+I
I......
Samplíng instants
-.I
T.
11.2 Voltage Source PWM Thyristor Converter
cussed later. The switching sequence pertaining to the example in Fig. 11. 7 b is plotted in Fig. 11.8, depicting the potentials of the motor terminaIs. Eq. (11.2) describes an idealised situation, where the protective intervals and the inherent delays of the switching devices are neglected. For the actual design of modulators, these effects must be taken into account, particularly the differences between tum - On and tum - Off times, which can cause con siderable distortion of the converter characteristics at low output volt age and frequency. Hardware- as well as software- based methods have been proposed for solving this problem [K35,S81,WlO].
of current
controllerTO
O. ,:",:---.
I
I
.
to I- : -!to I-! I ~ t2 ~ I
I
I
I/S1 ~S2
-1 . -
_.
I
I
I
I
I
.... _.
.-=.-:-1.._.J
º1~
~
1
1!:21
t
I
\A
._.-.
0_' _
º I ~21 ~ Iº1
1
Voltage vectors
11.2 Volt age Source PWM Thyristor Converter
Convener switching period 2T
o
I."ig. 11.8. Vectorial PWM with symmetrical switching sequence
Solving for tI, t 2 results in tI
1
USaRe!
USbRe!
To -
u;;- - v'3-U;;
t2 To =
v'3-U;;
2
~ ln = 1 _
USbRe!
USaRe! _
(11.3)
~
USbRe!
v'3
UD
'/:1
UD
> O. -
WlwlI ()lIlittillg zerO vectors, to = O, the resulting equivalent voltage vec 1."1 l'II
'//': I,,"f
= (1lI - 1l2)
tI
To
+ 112
;
(11.4)
= O and tI = To confirm this resulto Hence, by inclusion of dcsired voltage vector inside the hexagon defined by the six switching v('c/.ors lIlay be realised . Of course, the adjoining switching vectors must 1)(' choscn according to the sectorial position of 1ls ReC(t). If the switching rn'(/lI(~llcy io is high enough, fluctuations of the link volt age may be taken ill/.n accoullt by substituting a recent measurement of UD(t) for UD in the 1l1al.i.llg equations (11.3). T() relllove the ambiguity with respect to which of the two adjoining Hwil.chillg sUItes should be applied first, it is advantagcous to repeat tbe H11,11 1(' switchillg cyck with rcverse sequencc 50 that a syllundrÍCal switchilll.'; p;d.1.c1'll of kll/'. !.h 2'/'0 is ('.n~at(~d. Salllplillg l.ll(~ 1I10!.O!' cllrrcIlI.!-i, ill tll<~ C<~Jlt.(:r "r 1.11(' :i!im!. rircllil. illl.( 'rvlIlcl l.(I dilllilli::!iI'M1.111' n fl,llIpJill,~ lIoi:·;(" ;u, willlw di:.; !.II<' lilllitói for tI
l(J ally
I))()I!
225
The use of transistors (MOSFET, Bipolar) is presently restricted to lower and medium power applications, with lGBT's up to about 1 MW. With thyristors and GTO's, the rating of semiconductor converters extends to even higher power. A proven circuit, resembling the transistor converter in Fig. 11.3 is the thyristor converter shown in Fig. 11.9. Because of the constant link volt age and the low AC impedance of the DC link it is also called a volt age source converter. ln view of the higher power and the fact that the motor may have to operate for longer periods of time in the generating region, the line-side converter is usually an active front-end rectifier, able to feed power back to the line by reverse direct current ID. When employing gate tum off (GTO) thyristors, which are available in kV, kA- ratings, the converter circuit corresponds directly to the one in Fig. 11.3, with the transistors replaced by GTO's but with normal thyristors that cannot be blocked by gate signals, additional components are needed for implementing forced commutation of the machine-side converter; this is so because the stator windings of an induction motor do not contain current independent volt age sources as in the case of a line-connected transformer or a synchronous machine with an internal induced voltage. Hence, natural commutation of the machine-side converter feeding an induction motor is not feasible. This is understood when remembering that a converter with delayed firing calls for lagging reactive power which a normal induction machine is unable to supply. The converter depicted in Fig. 11.9 is an example of a variety of different circuitsj it is considerably simplified, showing only the main components. Instead of a pair of transistors as in Fig. 11.3, each phase contains a force commutated thyristor switch, comprising two main thyristors T, T', with ant.iparallel diodes D, D', two auxiliary thyristors AT, AT', a commutation capacitor C and an air cored inductor L. The resonant circuits in Fig. 11.9 have the ti\.sk of creating zero currentjnegative voltage intervals for the main Lby risl.o/'s '/',1" hy lirillg allxiliary thyrislors Al', AT'; briefly, a commutating JlI'()('I': ic: l.iI k" !i pla(', ' ;I: i f'()llows:
226
11. Power Supplies for Adjustable Speed AC Drives
Assume that the main thyristor T initially carries the load current iSl and that the commutating capacitor C is positively charged, Ue > Oj then firing of the auxiliary thyristor AT causes T to be blocked so that the load current flows temporarily through AT and C thus recharging the capacitor. At the sarne time a resonant circuit is formed consisting of AT, C, L and the diode Dj when about a half period of the oscillation has elapsed and the charging current tends to reverse its sign, AT and D are blocked, which leaves the capacitor with opposite voltage, ready for the next commutating transient when TI has to be extinguished. The load current is now fiowing through DI and the motor terminal 1 is connected to the lower DC bus. If the current pulse through the resonant circuit is of sufficient magnitude as compared to the load current, the commutation is almost independent of the load so that the voltage UlN can be approximated by a rectangular wave ±UD/2 with short but continuous commutating transients. The load current is l , Supply
----.--Generator \ Uo
Üo
Motor
-------r-----.. 0 Machine-side converter
iO
o
uo Inverter Rectifier Line-side converter
AT T AT' T'
lU 7N
Pulse width modulator Fig. 11.9. Thyristor converter with constant DC-link voltage (Mc Murray circuit)
hving co utinuous due to the motor leakage reactance, can in principie flow in citJwr dimction , even though alI four thyristors may be blocked temporarily, hCC
:wd
P "W l ' l
lI"w .
11.2 Voltage Source PWM Thyristor Converter
227
It is realised that the initial transient, when iS l is commutated from T to AT, is of course also continuous because smalI inductances, not shown in Fig. 11.9, limit the rate of change of the thyristor currents. Likewise there are paralIel RC snubber circuits connected to each thyristor to limit the rate of change of the voltages. A detailed analysis of converter circuits, taking alI these effects into account, is best performed by digital simulatioll'j this is particularly true for transient operation, when the initial conditions, for instance the charge on the capacitor when AT is fired, are changing. As with alI converter circuits, the function of the converter is eventualIy endangered when the required recovery time of the thyristors cannot be maintained, for example due to excessive load current or insuilicient initial charge on the commutating capacitor. Rence the accurate analysis of the commutating transients is of major concem to the circuit designer. Special types of thyris tors with short recovery time « 20f,Ls) are usualIy specified for pulse-width modulated converters, which limits the power rating to some MW. AIso, the occurrence of unacceptable load currents must be prevented by fast control action. The firing instants for the main- and auxiliary thyristors are determined by volt age and/or current controlIers. ln view of the limited switching fre quency, which is usualIy below 1 kRz for thyristor converters of higher power, the load currents can no longer be considered close to sinusoidal as was pos sible with transistor converters operating at a higher switching frequency. Rence PWM- thyristor converters of the type shown in Fig. 11.9 are nat uralIy suited to act as controlIable voltage sources, presenting to the load pulse-width modulated rectangular voltages, the fundamental components of which are prescribed by volt age command signals. Because of the large superimposed harmonics, caused by the chosen switching strategy, closed loop control of the fundamental currents at variable frequency is normalIy not practicalj it was different with transistor converters, where the switch ing frequency is higher and the currents are suiliciently filtered by the load impedances. Therefore, open loop voltage control through a pulse-width mod ulator is the standard solution, while the current loops are normally closed on the next higher levei of drive control, as will be shown in Chapo 12, The pulse-width modulators may again be of a variety of designs. One frequently used scheme with converters of higher power (Fig. 11.10) employs a triangular sampling signal [J8, M42, S29], the frequency to of which is a multiple n of the stator fundamental frequency ft to avoid subharmonics; n is reduced in stages as the stator frequency and volt age rise, to limit the switching frequency and to fulIy utilise the voltage capability ofthe converter. A drawback of this otherwise simple scheme is the appearance of large har monics ai the frequencies to ± 2ft . The load currents are smoothed by the leakage illductance of the motor which should not be chosen too smalI with this tyP(' 01' COllv(·rl.(\r. Ali tripkll harlllollic; a.re climinated from the cur l'('uLH iII :; 1., ':' 01.1' :: l.:t.I.,. dup 1.0 I.lw i :;ola.t., :d IlPIlLral
228
11. Power Supplies for Adjustable Speed AC Drives
are still noticeable additionallosses as well as sometimes objectionable noise cmitted by the converter and the motor. A typical current waveform with a PWM-converter and this type of "synchronous modulation" is shown in Fig. 11.12 b.
11.2 Voltage Source PWM Thyristor Converter Ü l1
n
2
UO
m=1 m=2 Numberof reversals per quarter-period m=4
Ar
v
t
I UI Ref
r;-;::::::-,
T
----.ó---.i
229
!1T> Tc
r
I' ,_o I
!
AT
u UI [Jo
m= 12
2
a
I
.
I
U
,,, ()II "
I I
I,
-1:1
I I
, I
I,
(1:) 1:2
1:
3
1:4
Voltage for m = 6
b
"
"\
1° Ttl
IN
1:
5
1:6
*
Uo -2
n 2 n 2
"wt=1: 1
W
1
t = 1:
Fig. 11.11. Optimised pulse-width modulation with precalculated switching patterns
/ "1
Triangular clock signal v(t) I"j,~. 11.10. Pulse-width modulator with a triangular sampling signal
' I'his unsatisfactory situation has given rise to the development of a mul f.if.II
J
allow the commutation to be completed. AIso, the losses in the converter caused by each commutation transient should be considered, which means that there is an upper limit for the switching frequency. At the maximum fundamental frequency, which corresponds to full output voltage of the con verter when feeding an AC motor, the optimal volt age waveform is a square wave with one reversal per half periodj this is included in the switching pat tern shown in Fig. 11.11. When realising a suitable modulator it is of course desirable to keep the required volume of memory smallj this can be achieved by storing only one quarter cycle O ~ T ~ ~ of the binary pattern for one phase and obtaining t.lte reruaining information by transposing and inverting this pattern. On the oLheI' hand , consi
(Ir 2 '0
lO )" ! HI.f'\IH pt '" qll adr' lllIl., t.hiH l ·t·llII ll.~ iII a IllclIlory
(Ir
11i
·.no
11. Power Supplies for Adjustable Speed AC Drives
11.2 Voltage Source PWM Thyristor Converter
II I! yt.( ~ which could easily be accommodated by one microelectronic chip of 11':1<1 only memory (ROM). At the very low end of the frequency range some (ll.bel' means of modulation must be employed because even 4 000 steps of :1.lIg111ar resolution per period will eventually prove to be insufficient. ln order to avoid continued fluctuations of the volt age amplitude, which would disturb the switching sequences and render the notion of "harmonics" 1IJ(~allingless, the superimposed control should contain a hysteresis band in 1.11(' amplitude channel; fine and rapid speed control are still possible through i.l1(~ frequency input of the modulator but the voltage would be changed more ~ :Imvly and temporarily in somewhat coarser steps. Some results of the current w:Lvdorms and spectra obtained with different modulators are shown in Fig. 1 1.12. The Curves were recorded on a 22 kw drive supplied by a voltage source I.hyristor converter [P44].
The load-independent commutation, resulting in low impedance motor volt ages makes the PWM converter an excellent choice for drives with high dynamic performance at power ratings where high frequency switching is not feasible; an example are high power traction drives with GTO-converters which require good dynamic performance to damp oscillations in the drive train and prevent wheel slippage during starting and braking.
1~L~ ~
+jQL
ULI
iLl
-I 11111 III 11111 IRlm IL;
U
~ 11111 Wulli±illu~ III11 ~U~
IN
231
machine-side PWM converter
•
t
Uo
iSI Une-side PWM converter
~II~I~I , I __~I~L\~~~M~o___
iSlv
17
29
;/
__ I
1
fifI
b
13, 17
29
43
~, fifI
I>'il-(_ 11.12_ Current waveforms and spectra achieved with different methods of I'lIls(· , widl.h modulation. (a) Optimised switching patternj (") SI! hharmonic modulation with triangular sampling signal
fi particular feature of the pulse-width modulator with optimised switch
pattcrn is the fact that the pattern can easily be changed for different :1 pplical:iolls, placing more emphasis for instance on losses, torque pulsations
' ·"IIVCI'I.('l'ii
111(iIJ , II(i71 .
Fig. 11.13. Symmetrical voltage source GTO-converter with forced commutation at machine- and line-side
With increased use of controlled drives the problem of line-side interfer ence by power electronic converters is becoming an important issue for the utilities. A line-commutated converter supplying the DC-link is particularly notorious because it not only produces distorted line currents with low orders of harmonics but draws also substantial reactive currents of line frequency. The first mentioned effect can be alleviated by increasing the pulse number of the converter but the other requires capacitive compensation or, combining both aspects, filters with capacitive fundamental input currents. Another possibility that is likely to be of strong future interest, is to extend the principie of forced commutation also to the line-side converter. An example is seen in Fig. 11.13 where a PWM volt age source converter with GTO-thyristol's at both ends is shown in simplified formo The line-side PWM t;OI1VÜl't.t:J' c.olllrl llC cOl1trolled ::;ut:h that the line currents iN are not only a ppJ'()a('hill i\ Nillll li()ldld Wav({oflu ImL LhaL Lheir fundamental components are n.I I:" iII 1'1111.: ;1' \\' ll.h !.II<' lillC' v()ILa.l'i'H, LlIIIH t,jilllill fl t.ill ({ l't'aC'l.ivu CIIJTtmLR. Singll'
'2:\2
11.3 Current Source Thyristor Converters
11. Power Supplies for Adjustable Speed AC Drives
233
Supp/y
phase converters of this type are common on large AC-traction drives, where 1.11(: line impedance caused by the single phase catenary is quite large so that voltage fluctuations and distortions would be particularly severe, Fig. 9.17. The line side converter may also be used as a static compensator by producing controllable, usually capacitive reactive line current, whereas the ;I.cl.ive component of the line current serves for maintaining a commanded ('OlIs t.ant or variable link volt age UDj this is discussed in Sect. 13.2.
LO
iNl
io
Ts ~---lll
t
omitted with GTO Os
Uo
I 1.3
°2
Current Source Thyristor Converters
'1'1((: converter circuit shown in Fig. 11.14 comprises once again a line-side
11IId a machine-side converter part connected through a DC-linkj the two co"verters operating at different frequencies are now decoupled by a smooth illg reactor L D which, in combination with a current control loop, serves Lo lllaintain a direct link current iD as prescribed by the current refer "IlCC. Thus the machine-side converter is effectively supplied from a current SOllrce [F2, K56, P32]. The circuit offers the following simplifications: • The unidirectional link current iD allows the use of a two-quadrant line (,()l\verter, where reverse power flow is achieved by inverting the mean of 1.11(: link volt age UD through delayed firing. • 'I'hc link reactor permits the volt age UD to be temporarily raised or low crcd during commutation of the machine-side converter, thus avoiding the ;wxiliary thyristorsj commutation is now performed with capacitors and d("'ollpling diodes. • 1\ II., 'I;d.her there are 12 thyristors, the sarne number as would be needed for II. l'l'vcrsible DC-drivej also, the thyristors for the machine-side converter I" '"d llot be of the more expensive fast recovery type necessary for voltage S'lIlrcc converters. Tire DC-link could also be supplied from a force commutated converter :;llch as a chopper with current controlj this is of interest on subway trains f('d w i til constant direct voltage. 'I'lw lllode of operation of the machine-side converter in Fig. 11.14 differs "ollsiderably from that of the volt age source converter, Fig. 11.9. While the 1I",Lo!' cmrent supplied from a PWM voltage saurce converter was at least I'I'lId<'iy sillusoidal, it has now a three step waveform, assuming the values 'i/l, O, ij) as seen in Fig. 11.15 a; the corners are rounded-off due to finite (,olllllllll.alillg time. The vector of the stator current follows the trajectory of ;1. :: 1" poillted star; this is shown in Fig. 11.15 b which was recorded duriug a ::J",, 'd l'I 'v(,rsal of a. 22 kW drive [G5]. T"(~ cOllllllllI.aLillg t.ransiCllts
T2
iO
is!
iS2
iS3
iORef
a Uo
Motor io Generator
b Fig. 11.14. Thyristor converter with direct current link (a) Circuit (b) Operating regions of DC link
The commutating transient may be briefly described as follows: With no commutation in progress, two thyristors, for example TI and T 2 carry the direct current, iSl = -iS3 = iD, and the capacitor C l3 is positively charged as a result of the previous commutation, UCl3 > O. If thyristor T 3 is now fired, T l is extinguished in a rapid transient and T 3 assumes the direct currentj this is the starting condition of the commutating transient proper. While the current iS l is now reduced towards zero, iS2 is rising towards iDj during this interval phase 1 of the motor is fed through C l3 as well as through the series-connected capacitors C 35 , C5l , Eventually diode DI is blocked and the commutation is completed with T 2 and T3 conducting, UCl3 < O, UC35 > O. Clcarly the motor transient impedance between terminal 1 and 2 is part of lhe CQmmutation circuit, as mentioned before. The diodes are required for dtcol.lplillg t,o prcvent the capacitors from losing their charge, necessary for L11(~ lt( ! :{(. cOlllllllll.al.ioll. The cOlI\lHlltatill['; inl.(~rval call be reduced by choosing It IlI o l,o!' w i I.lt low l<'ll k:tr~e n ,a,cl.allcl: .
U I II l 1 (, I I
I' f\
1
234
11.3 Current Source Thyristor Converters
11. Power Supplies for Adjustable Speed AC Drives
on the slow progression of the state of the counter; this is discussed in Sect. 12.4. Of course, due to the relatively slow commutation at light load, the switching frequency is bound to be much lower than with a voltage source PWM converter [W9]. It is for this reason that this converter is not suited for drives requiring continuous position control such as machine tool feed driveso Apart from this special restriction, the current-source converter may be controlled nearly as rapidly as a voltage source converter, because the firing instants of the machine-side converter can be quickly shifted; changing the magnitude of the link current involves some delay due to the large smoothing reactor but with adequate ceiling volt age the equivalent lag can be reduced to about 10 ms with a 50 Hz supply, as was discussed in Sect. 8.5.
15{t}
\,/
...
-~\
I a
235
o
b
Fig. 11.15. Current and volt age waveforms of a current source converter (li) Wave forms; (b) Stator current vector
Beginning and end of the current steps are manifested by volt age spikes superimposed on the otherwise sinusoidal terminal voltages of the motor; this ix (ln obvious consequence of the changing stator currents, Outside commu l.aLiOIl, where dis l/ dt
= O,
IJ
= 1,2,3
li d.n l.l rt[', I.h"
d lll ', ( '( IIIIlI.('I ' ,
LO
io
(11.5)
h()lds, Lhe terminal volt age of the motor is practically sinusoidal since the III;ll~lId.i s illg current is filtered by the main time constant of the motor. IkpclIding on the phase of the voltage surges due to commutation, as rd;i.I,cd to the sinusoidal volt age component, it can be recognised whether the ""I,('hill(~ is operating as a motor or in the generating region, Fig. 11.15 aj on 1.1((' 1)( : si de this is reflected by the sign of the mean volt age UD. The direction (II' \,o(.a(.ion is solely determined by the firing sequence of the machine-side (,()lIvcrt.cr. The firing signals may for instance be derived from a ring counter wil.li (j states, which is stepped forward or backwards as commanded by the :;I)('c
G" ,
Machine-side PWM converter
L (' , :illi'('r! lIq,u fil rllfo / ~ II n.it."r'Il l1. l.il ll (
' · (lIll1)(lIIC'oIlI.
Uo
Pwu,;~;~:rle,
'1
'
D
· :.:
II~, l :I~ -
'"
ue'
~ /~
,,'
US11 Fig. 11.16. Symmetrical current source PWM converter with GTO-thyristors
As mentioned before, the current vector defined in Eq. (10.5) assumes wilh this converter a rather unique shape; because of isv = ±iD and with 0\1<' of 1 ,lIc rnotor currents zero (except for commutation intervals), the vector roHows a rq';\I1ar six-pointed star with thc [adiu::; V3iD, ·i ,<: (t.)
': :: 1
1::; '. (..i
'i s :1 (':i Z
J:\-í /l
( oJ
(I ' '( / 'h I rr /Ii)
(lUi)
~n6
11.4 Converter Without DC Link (Cycloconverter)
11. Power Supplies for Adjustable Speed AC Drives
Some measured traces are shown in Sect. 12.4. When assessing the relative merits of the different converter schemes, it is of course necessary to compare not only the number of components but ;Liso their rating and how this is reflected in the total cost. The circuitry of the current source converter is again considerably sim plilied when GTO-thyristors are employed; this is indicated in Fig. 11.14 :lIId shown more clearly in Fig. 11.16 for a symmetrical converter, where the I. ';1.<\- and the line- currents are pulse-width-modulated and may approach :,i 1\ IIsoidal waveforms. The link current iD is now switched between the motor terminais and the 1II(,Lor voltages and currents are filtered by capacitors. It is also possible to i IlLroduce zero current vectors, bypassing the load through a short circuited I(~g of the converter which is similar to the zero volt age vector with a voltage Silllrce converterj thus, the circuit offers possibilities for high dynamic per ronllance drives. However, in view of the increased complexity of the control pl:\ut, caused by the capacitors, sophisticated control algorithms and a pow nflll microcomputer are neededj in particular, the question of L-C resonances J( ~ qllires detailed study [A18, A19, N17, N18, N19, N21].
By supplying each phase of the motor winding from a reversible converter, a low frequency AC drive system is formed, as seen in Fig. 11.17. Again, the neutral of the stator winding remains isolated to exclude zero sequence, such as triplen harmonic, currents, The main feature of this circuit is that only standard line-commutated thyristor-converters are required which have been in use for many years with DC drives and HVDC systems up to the highest power ratings. AIso, the cost of the thyristors is reasonable since no particular specifications with regard to turn-off time are necessary with line-commutated converters. Netz
- - - - - -
~1~
Neutral 01 converter
I " I " I
II
/
Converter\ IConverter phase 2
11. 4 Converter Without DC Link (Cycloconverter) TIi(~
237
phase 3
UIN
(:ollverters discussed so far had in common that line-side and machine
::id(' ("()llvcrters operated at different frequencies and were decoupled by a 1)(: lil\k colltaining energy storage devices in the form of low-pass filter com
1")\I(')II,s. 'l'his twofold conversion provides a degree of independence in con l."I)lIilll': Lbe converter ends, but at the sarne time causes additional power I",,::.,::, Til,' double conversion may be avoided with direct AC/AC-converters wlii('1i I'crforrn the conversion from the constant frequency, constant line volt ;"'," I,() variaLle frequency, variable motor voltage in one stage and without 1I'::ml.illg to intermediate energy storage [49,58]. TJlI ~ rcversible line-commutated AC/DC converter without circulating ('III\'(')lt shown in Fig. 9.4 is capable of operating in ali 4 quadrants of the '/I,,,I'/. .. pl ane supplying an inductive load (to assure continuous current flow). I kll(,(~, by providing it with closed loop control of the output current and il.pply illg (l low frequency current reference, the converter will act as a low i'\'(''lIJ(~ll('.y current source. Clearly, 4-quadrant capability is necessary when ::lIl'plyillf'; a motor because, when a sinusoidal current ia is fed to a R-L I,,;ul illlpedance, the operating point in the voltage/current-plane describes ali nllíplical path around the origin, passing tluough ali quadrants. When lI 'v( 'l'iii)l!~ t-he sequellce of the cnrrent rderell(,(~s, t;Jw sequcncc of the thre(~ pll;I.:;(' Olltpllt syst(~lll is inverteu. To avoi<\ IIl1d(':;ir;Lhi<' iJlt(~ractiI)IIR UelW( '(, 1l 1.11(' 1.111'1 '<.' <.'111'1'('111. ('OIlI,l'IllI.'I'S (OIH-~ (J[ wiIich i: ; /'('dlllldil,ll!. III:(',n,ll.s(~ of UI(' i',(,ro :111111 "i' 1.1 1\" ("111'1'/'111.:; ), i.lwy IlIlI.y 1)(' r11":i(,\1 ,.,1I' ili li l./'illl pr 1'1 (,oJlI,roll('rn I,hwi oI llI wiul'. Ilfl 1II" IJl'Nil,il li,\,
iS!
tiS!
_ _ ..J
iS! Ref
Fig. 11.17. Six-pulse cycloconverter for a three-phase motor load
On the other hand, the large number of thyristors seems at first sight staggeringj for a six-pulse converter with three-phase output a minimum of 36 thyristor branches is required as well as a transformer with three complete sets of three-phase secondary windings. This indicates that cycloconverters are mainly of interest for large drives, where parallel thyristor branches would I>e necessary with other converter circuits. An important restriction is the li1l1itation of output frequency which is caused by the discrete nature of the collirol process and the presence of a carrier frequency, since the output volt af~e:) ar( ' :tss(~lllhlc(1 frorn scctiOllS of thc lin(~ volta.[1;(~s. As the output frequency l'i s(,:-i , t.lw olli.pll!. CIlITI~llt.S ;tI'C t.ra('.killl~ t,h t' sinllsoidal rderenees with incn~ as illl~ "' ITo m fl. 11I1 ('(JIIS('qll('n!. di H(.OItjoll , 'f'11I' freqll('II('Y l' a ll I ',I ,
:J.:lK
11. Power Supplies for Adjustable Speed AC Drives
O :::::
ph
h : : : hmax ~ 15
11.4 Converter Without DC Link (Cycloconverter)
(11. 7)
is usually considered for operation of a cycloconverter, where h is the line frequency and p is the pulse number. With a 50 Hz grid and a three-phase bridge circuit (p = 6) this results in h ma", ~ 20 Hz. At this frequency, a period of the output volt age would on the average consists of 15 sections of line voltages. Of course, if a three-phase supply of higher frequency is available, the range of output frequency is extended accordinglYi this may lJe the case on vehicles, aircraft or ships, when a diesel- or turbine-driven gcnerator provides on-board power for variable frequency AC drives. AIso, a somewhat wider frequency range could be achieved by cycloconverters with circulating current, as discussed in Sect. 9.1 [H28] .
~~~~~
_
ult)
jjH~~~
~'02
_
jj~~~1
~ml'1I11.1m"'1lm~_rr", ; H1U
iR~/i(t)
/
MHII,
~
"\.J
'C7
v
/
I
1\ f\ (\ 'V V \J V;
~ef(\
i(t)
'---:l
/
239
/
/
Fig. 11.18. Waveforms of six-pulse cycloconverter supplying single phase-load at different output frequencies (simulation)
ii,
Tt is of interest to note that the nature of the load supplied by the cyclo COllverter is unimportant as long as it contains a sufficiently large inductive illl(w(lance to assure continuous current flow. ln particular, there is no differ ellC(: whd.her til<: load is active or passive, i.c. wltcthcr thc HlachiIl(~ operat(:s ii:;
II,
11.l'IIIIIII"(,r ic'lI ll .hl 'C'"
1'111111" 11 .1': :1."111
"r l4 i llll ~l"i dHI
Fig. 11.19. Measured output current and volt age of a six-pulse cycloconverter with three-phase inductive load at 10 Hz
volt ages and currents results in constant total power. Since the cycloconverter contains only switches but no storage devi ces (apart from the unavoidable leakage inductances, snubber circuits, etc.), the total three-phase input power corresponds to the output power. Hence it is mainly the effects of harmonics that have to be considered. ln addition there will be reactive power at the line side which is inherent in the control of line-commutated converters by delayed firing. ln Fig. 11.18 the results of a computer simulation are shown, where a reversible six-pulse converter with current control, operating on the 50 Hz grid, supplies a predominantly inductive single-phase impedance with cur rent of 10 Hz and 20 Hz. At these frequencies there is still a reasonable agreement between reference current and feedback signal, even though some phase shift becomes noticeable in the 20 Hz-trace, but it might be corrected by a lead-term in the reference channel. The 2 ms-zero current interval, which is necessary for safe operation when activating the opposite converter bridge, is clearly seen. With a three-phase output and symmetricalload, all triplen harmonics would be removed from the load current. Due to the inductive load the volt age rises with increasing frequency. This will eventually lead to addi tional distortion of the volt ages and currents caused by the ceiling voltage of the converters. Finally in Fig. 11.19 oscillograms of the measured output current and tine-to-neutral volt age of a cycloconverter are shownj the converter is supplied from the 50 Hz three-phase line and feeds a symmetrical three-phase inductive load at 10 Hz [AI3]. For large drives, synchronous motors are often preferred ove r induction motors since there the reactive power does not have to be supplied through I.!te converter and because a larger airgap is desirable for mechanical reasons. Cydoconverters are very suitable for this duty, provided the stator frequency is s ufficicntly low, which is often the case with large low speed motors such a s rcquircd for rolling mill and mine hoist drivesj synchronous drives will be disc lIssed iII Clw.p. 14. Ás : \.11 a.11.(:rruüiV(' t,o DC Iink PWM converters, force commutated cy C"\O('OIlV(' I' I.\'I'fI , (".I dkcl HllI.l.rix eOllvert('J's f1'~R, S;J7], JIIay bccome attractive at lc/w l'\" !,"IV"\" 111,1.1 111'" 1"" '; 1.11:-... 1.1", ('III'I', (,Y :ll.o l' lI,l'," "'IJIIJl0J\('lIts in t,h e nC· li llk
:HO
11 . Power Supplies for Adjustable Speed AC Drives
12. Control of Induction Motor Drives
L
I I
c
Ul2
8 /1pply
2./' ~ I
-o
2
ó
U23
3.,,101 2
D
3
-o
t
3
Load Principie of a cross bar connection
b Example of switching arrangement
TC
'6 UI
u2
U3
U,2(1)
w,t = f c Selection of fine voltages I;'i~.
~[
tI
d Bidirectional switches
11.20. Principie of a PWM matrix converter for generating
viLri"tble frequency voltages directly from the line volt ages l U . ' a voi(I( ~d , thus facilitating large scale integration with semiconductors. TIl!' prillciple of a matrix converter is shown in Fig.11.20 a in the form of II. crel;;S h;1.1' al'rangement, where the three load terminals are alternatively (·"I!II!'cl.I 't1 1.0 the three supply terminals. This requires nine bidirectional Ii wÍl.dl\ ·:;, t"Íf.( .11.20 d, to be pulse-width-modulated at some kHzj a possible ;•• ' :\lII ',1' III('Il[. is seen in Fig. 11.20 b, where in each time interval Wl t = 1r/6 iiI<' Il il'; lwsl. tine-to-line voltage assumes the role of the former DC link volt ai '.• ' , ;I.~: shown in Fig. 11.20 c. This switching strategy permits the highest pO: ;: : lhl( ~ output voltagej at the sarne time the reactive line-side current is n'd u('( ~d sillce the current flows only in the centre region of the line volt 'q'." p(' rioc!s. The higher frequency components of the line currents can be l('lll()v(~d hy capacitive filters. As usual, it must be prevented that the line v()ll.;q ;l ~s a re temporarily short circuited but this is excluded with the arrange 1II I: lll. of sw itcbes in Fig. 11.20 b. The equivalence for vectorial modulation \V Iii.:" w rrcsponds to Eq. (11.2), then reads
'/t s I!.,.r
= (USa + j USb)Rer = U i k (r) ( ~~)
_e j ,,/3
12) 1'0 '
(11.8)
wll( ~ l'! ~ 'IIid'1') is t.lw hil~ll cst 1i\H ~" I,()· lin(' voll.lIl ~'· iII ('jL('h 'Ir j (j .. illtnva.l. ln adcli· tillll 1'1 I I..... I 2 /'0 'III Ilolds, wlwn ~ 1'0 ii-! 1.110' '1. ('1'0 vI'cl,or i lll.<;rval , wbclI I.hc 11I1I!.!,r i: : . . h"rI. dl'('uil.l ·d, a.1. 0111' 01' 1.11(' lill!' 1.(·I·II IÍ llill rI, f i/\, II)!.
When comparing the dynamic model of a separately excited DC machine, Eqs. (5.1-5.4), Fig. 5.4, with that of an AC induction machine, Eqs, (10.50 10.53) , Fig. 10.16, it is obvious that the latter represents a much more com plex cOntrol planto This is caused by the fact that the main flux and the armature current distribution of a DC machine are fixed in space and can be directly and independently controlled while with an AC machine these quan tities are strongly interacting and move with respect to the stator, the rotor as well as each otherj they are determined by the instantaneous values of t,he stator currents, two of which represent independent control variables. An additional complication stems from the fact that the rotor currents cannot ue measured with ordinary cage rotors . Hence the AC motor is a nonlinear multi-variable control plant that kept cOntrol engineers puzzling for a long time. The differences in control dynamics of a DC and an AC motor are best <~xplained by the mechanical mo deIs shown in Fig. 12.1. ln Fig. 12.1 a, cor I'csponding to a DC motor with a mechanical load, a disk is driven by the I.angential force Ir acting on a pin P which can be moved in a radial slot I,y a radial force IR acting against a springj caused by velocity-dependent I'ri ction, the radial motion of the pin is assumed to be relatively slow. Between this mechanical arrangement and the electric model of a DC II I
IT
~
R
~
IR
~
armature current, main fiux, field voltage,
Rh RW
~
R W IT
~
~
electrical torque, induced voltage (e.m.f.), electrical power.
Controlling torque, speed and angular position of the disk is straight IllI'w a rd if h and IR can be independently chosenj the analogy applies to "pI'I'atioll below base speed as well as in the field-weakening range with the lill,il.a l.ioll bcing either saturation (length of the radial slot) or maximum ·'I ,dll ('.(~d voJtage (circumferential velo city of pin). 1\ ::illlil ar arrangement a pplics to the AC motor, Fig.12.1 bj however, 1. 1J l' pi ll i~; 1I0W drivcll by three COllllecting rods that apply forces h , h, ,I III l.il l'I'I' li x(·d din'l'.liollS Rpa.<:ed hy '120° . To procluce a smooth circular III,tI.io ll wil il,' 11.1. I.II!' /lalIlC' t.il1w 1
242
12.1 Control of Induction Motor
12. Control of Induction Motor Drives
'2
'R
243
stator resistance and assuming symmetrical sinusoidal stator voltages UI of frequency WI , the electric torque of the induction motor is given by Eq.(10.85) where the normalised slip derived from Eqs.(1O.67), (10.71) is
I'T res
S Sp
(wl-w)aLR=W2 aTR.
(12.1)
RR
By varying the stator voltages in proportion with frequency, UI Uso Fig. 12.1. Mechanical model of (a) DC and (b) AC motor
WI
(12.2)
Wo
the stator flux is kept constant, hence the steady state torque is
2 lc for DC motors; instead, a large number of contending proposals existed, 1I10st of which were only useful for special cases or particular applications. IT()w~~ver, since the beginning of the 70's, new methods adapted to vectorial Illachine mo deIs have been proposed that had the potential of forming the ba :;is for a unified theory of controlling AC machines. ln the following chapters, HOIIII ' ()I' Lhese methods will be discussed.
mM=mpO S
S -sr> + ~ S
=2m po
W2 aTR T)2 ( 1 + w2a R
The torquejspeed curves are obtained by shifting this reference curve along the frequency axis according to W = WI - W2, while maintaining its shape; this is seen in Fig. 12.2.
ül
rol
mM
ContraI of Induction Motor Bll S(~d on Steady State Machine Model 12 . 1
s = consto Sp
IlIilIly mechanicalloads, such as pumps or fans, there is no need for high dyll;ullic performance, as long as the speed can be varied with good efficiency ()v( ~ r the desired speed range; this permits the use of steady state machine Illodc!s for the design of the drive control. Two of the schemes that have been pl"oJloscd for this purpose and are in frequent use, wiU be discussed first. They il.r<~ based on the assumption that the induction motor is supplied by a PWM voltuge source converter, producing symmetrical three- phase voltages and CllrrClIts of a fundamental frequency related to the desired speed. III vicw of the simple and very effective control schemes for DC motors it w;\,~; appr()pria(.(~ to sca.rch for similar methods that might be applicahk to I.h(~ (·.oIIt.rol o[ .l\.C motors. A salient fcatllr(~ wit.1J De 1II0tor coutrol wa,'3 to 1I11l.iIlt.nill rOIlHL;uIl. Ilux l)(~I()w hase Hp('ud t.o fllll'y IIl.ilü'I! t1Hl g,iv()Il UloLor ~;i'/'(' 1'(lI
wiLh 11'1'.:1.1'<1 L
W;I:;
"oIlLr"lh'd I.hl
IlIiI)' I... I'
(12.3)
nl'lllil.l.lIl'f·
("1I1T'·.III.;
1,,1 I.h,· 111'''11 <1 111. IWI',lc'dílll':
a
iiI<'
- wQt--_--, ~
(J
~
= 0.2
=0.1
Rs=O
Fig. 12.2. Steady state torque/speed curves at different stator frequencies, ;lssuming constant stator flux and neglecting stator resistance
If only the centre portion of the curves having negative slope is used, i. l>y limiting the rotor frequency to values below pull-out slip, the curves ('xhibit a marked resemblance to the steady state torquejspeed curves of a I )C- lIlachine with constant flux, Fig. 5.5 a); hence, the rotor frequency may !,( ! rq~;tr( l ed a.s n ~pre:;cnting torque, scrving as a substitute for the armature
(!.
CIIIT(!III.I.l1 7,U:III · A H ll lllf'. 11:1 1.11<' v()ll.al'/·
12. Control of lnduction Motor Drives
'244
12.1 Control of Induction Motor
occur at low speed, where the resistive voltage component will eventually (;xceed the induced voltagej hence, a correction of the stator volt age according 1.0 Fig. 12.3 a is required,
Ij::
Us =
Uso
+ Rs Isl ;
(12.4)
ir the operating range is to be extended to low speed . With Eq. (10.72) this rl;sults in
-I
Us -
Wl Wo
I
Rs 1 + j W2 TR Uso· + j Wo Ls 1 + j W2 U TR
USo
I
245
....
USI Rerl
{g
I U~O D",I
oc: u
~
r:-<
= ,,'
100
001
Controllaw
Q)
2
U,lIf'-./
Wo
00 1 Re:
I
I I
I
US
3Re ',
oU)=> Q)
O>
.l!!
~
00
(12.5) 00, = O
a
~ Uso 1.0
Fig. 12.4. Speed control of an induction motor with impressed stator voltages and open loop flux control, based on a steady state model of the machine
R
.:.:s.
~~
= 0.01 , Sp = 0.2 / h/
SIS p = 0.5 ~
IS
UsLa
Rs
uRLa
IR ~
0.5 U= 1
~U 01 SJ
U;.;
0
Lo. 1m
'/
~:"
'/ '/
'/
'/
'/
~
~
~
'l
eliminate the effects of stator resistance and the primary leakage reactancej this would also simplify the dynamics of the drive by removing Eq. (10.50) from the model equations. ln addition, the assumption of a sinusoidal supply is much more realistic with stator currents than with voltages produced by a switched power converter. As explained in Chapo 11, either a PWM volt age source converter or a cycloconverter with current control could be used, covering a wide range of power ratings. When supplying the motor with impressed sinusoidal currents, it is ad vantageous to aim for constant air-gap- or, better yet, rotor-flux, because this eliminates not only the effects of stator resistance but also of leakage inductancej in the equivalent circuit in Fig. 10.7 b, this corresponds to the condition of maintaining a constant rotor- based magnetising current ImR or rotor flux, independent of speed and load. The pertinent stator currents are obtained from the equivalent circuit, resulting in
'l
'l
'l
'"
'l
O O
1.0
0.5
b
n
001 00 0
Torque/speed curves of induction motor at variable stator frequencYj valcnt circuit, (b) stator voltage as a function of stator frequency .. o" s l.;\.II(. stator flux at S = Sp/2
1,' 11\ . 12.:1. (,,) " " II i
I,,,'
b
= 0.1. Devia the idealised Eq. (12.2) are noticeable mainly at low frequency and 1\1. w [ = O, the necessary volt age is caused only by the stator resistance, which dcpc llds on the winding temperature. ln Fig. 12.4 it is seen how this func· I.ioll cau be incorporated into a simple speed control scheme of an induction IIIOI.Of. The speed controller is producing a reference for the rotor frequency whidl is taken to represent torque; it must be limited to the range where 1.\1( ; I;orqllc/speed relationship is approximate\y linear, to avoid pull- out. By addillg LlI(: signa\ for mcasured speed, a referenec for lhe stator frequency w'j ii> ohl.aú wd . Naturally, sueh a simpl0. eontro\ ~dIemc is oJl\y ~mit ablc for spcl,d ;\djll stllH'1l1. iII stl ;ady st.atl·, not for Itil(h (>1:rfonll:lIlCI; drivr:s. 'l'lte curve in Fig. 12.3 b is calculated for Rs/wo Ls = 0.01, S
I.inll s
[rom
()\, I' II
~il L iv , \ 1. ..
Innp ('nllll W II S ll.t.inl.~ ("ollt.rol s("It' : I1I" ~ Iii\('
LlI<' lllll:
ili d l ' ,,11','1'1,11 !1II ,·1t ! tH 1.'·II'l)(' l'I d.III '" l 'IIiUlf ':" '1 1 Ll ll d, 111'(',
1\" 11 111'"' 1"' ''" ' ''1. IV"lIld ii" L, ) ('() !lV"
11.
de~i,(·. ril)( :d nn: /ln!.
n,1·(' () IlIII.( ·d
1'1 '(1111 \ lIlJli <'rIlIC' d v"ll. n,f',C'll 1.11
KI.: II
ror .
CII I' I" ' III :I
1. ..
jWl(I-U)LsS(I+UR)2] I -mR I = [1 + -s RR
=
(l+j w2 T R)I
mR
· (12.6)
To maintain a given rotor flux, the magnitude (RMS- value) of the stator currents must follow a function that depends on the rotor frequency W2 and contains only rotor parameters,
Is =
VI + (W2 TR)2 ImR .
(12.7)
This is achil'ved wit.h the help of another function generator which produces a l'C,rl ~ I'(·ltr' · a ccordill g t o E q. (1. 2.7) for the amplitude of the stator currents. TIl<' rlllll'i. illll i l4 plll!.I., ~ d iII [Cig. 1:J. .r. for I.wo Vil h)( ~s of UI(' rotor time eon stant; (l , L, ' r' l" (l p,n ;1I.l, H! IH ' I(,w /1.1111 n.l)()vI ' sYIldlr
12.1 Control of Induction Motor
12. Control of Induction Motor Drives
246
speed. Of course, the fact that the function depends on the rotor time con stant constitutes again a source of inaccuracy since the rotor resistance may vary considerably with temperature and the inductance with the degree of saturation. Even though this might be alleviated by making the function gen crator dependent on temperature, the realisation would be difficult in practice because a measurement of rotor temperature is not easy with the motor in operation.
~
'mR
247
Thís confirms the earlíer observatíon that the torque/speed curves may be shífted along the speed axis wíthout changing theír shape, Fig. 12.2. Since it is now the rotor flux ínstead of the stator flux that is maintained at constant value, the effect of stator leakage is eliminated, resulting in a reduced slope of the curves. The next task in the design of the AC motor control consists in converting the current amplitude reference Is Ref produced by the function generator to a system of three-phase variable frequency AC signals to be used as current references for the power converter and motor. This is a modulation process for which various realísations existj two possibilities will be discussed here, one analogue and the other digítal. x X2
Y1
Yt
JX7Ref
r+0t".~----
Y2
x
1.0
I~I
íG ~
2. . / ...--1
•
Y2
000
X,2
•
(J
X1
2./~
Cú
2
a
With the definition of the rotor- based magnetísíng current ín Fig. 10.7 b
I s + (1 + aR) IR
= 3 Lo 1m [Is I~] = 3 (1 -
Control/able Oscil/ator
a) Ls 1m [Is I;"R]'
(12.9)
t\ SS lluting a correctly tuned functíon generator, the magnetísíng current wíll hCll(:c 2 - I 50'
(12.10)
·..·m R-mR -
which ou insertion into Eq. (12.9) yields mM
= 3 (1 -
The block diagram of an amplitude-controlled, variable frequency two phase oscillator is shown in Fig. 12.6 a which is suitable for generating the required AC reference signals [H26]j similar structures were contained in Fig. 10.16. The input signals YI, Y2 and the output signals Xl, X2 are connected hy the following nonlinear differential equations
a) Ls I~o 1m[Is/ I mR ]·
(12.11.)
'0
:1 (I .' " /,:: 1 .~·tI é<.J ~ '1')0' . )
dt =
To
dt
dX2
-YI
X2
+ Y2 Xl
,
(12.13) =
YI Xl
+ Y2 X2
,,2
ct
2
:J;,
dXl
lo -.~ - 2 y"). T O- IlI ' dt
0
~/
dXI
To
.
loaimination of X2 results ín a second order equation
11'0 111 Iohi s foll()ws wi1.h Eq. (12.6) '//1
b
Fig. 12.6. Variable frequency, variable amplitude two-phase oscillator. (a) Block diagram, (b) Eigenvalues in s-plane
IlC lIIaintained at nomínal value, correspondíng to the ideal no-Ioad current,
[ r
Amplitude control
Y2
Y1
(12.8)
I.lw st. ( ~ady state torque accordíng to Eq. (10.81) may be written in the form mM
000
x
F ig. 12.5. Amplitude of stator currents required for constant rotor flux
l"'li =
CD;;
(I:U:l)
2 dYI + (2 YI+Y2)XI=-To dt
dY2 x2+TO-d t
Xl'
(12.14)
t\: 1:11 1111 illJ'; Y I, y ~ ("ou:;!.., a. lilll'ar hOlll01!,('!I( ~ OIlS differential equation is ob I.II.iIlC ·d, I.IIC' c·h fll'tI,f"I.•·, i,.Uc· c''lllal,ioll .,1' wl.i.-l. i: :
12.1 Control of Induction Motor
12. Control of Induction Motor Drives
248
Tg S2 - 2 Y2 To S + yi + y~
= O.
(12.15)
With Wo = l/To the two complex roots are S12 . - ' =Y2 ±JYl;
(12.16)
249
and 1;!CI~ef' Following a step change of Yl, there are no transients at ali as the oscillator simply continues from the existing initial conditions with the new frequency. The state vector ;!C (t) is a continuous function that is well suited for use as current reference, as it enables the current control loops to track the references with small dynamic errors.
Wo
t.he real and imaginary parts are subject to decoupled adjustment by Yl, Y2, as seen in Fig. 12.6 b. Yl = consto i- O and Y2 == O give rise to an undamped oscillation of fre (luency Wl = YlWO; changing the sign of Yl results in an inversion of the phase scquence. Correspondingly, the amplitude of the oscillation decays for Y2 < O iLud rises for Y2 > O. Since a constant amplitude I;!CI = xi + x§ is desired in stcady state, this calls for an amplitude control loop as is indicated in Fig. 12.6 a, where Y2 serves as actuating input. Even though the control loop is llonlinear, its stability is uncritical so that a PI-controller proves adequate; when entering the square of the amplitude reference, 1;!CI~ef' the root function ntll be omitted.
012 ReI
IS (0l2) Load angle
J
l~'- - ------.
1/ 1'< +4
V'
r-----
,----
J
I~~
b
c
h~~ ~
Ol
Ol
T~
oo:ib:Y
Fig. 12.8. Speed control scheme with impressed stator currents and open loop control of rotor flux
t~'_______ _ 1 --
I~-
sin C
The two-phase orthogonal signals Xl, X2 are now combined to form an equivalent three-phase system WI, W2, W3 to be used as current references
;!C(t)
= XI(t) + j
X2(t)
= %[wdt) + eh W2(t) + e-h W3(t)]
,
21f
,= 3
'
(12.17)
To render the transformation unique and exclude a zero sequence component, Y~ -- --'I ,, x,
Wl(t) + W2(t) + W3(t) = O
,
L ______ t
is added as a side condition. A linear combination of Xl, X2 results in a set of symmetrical three-phase reference signals for the stator currents,
t,
I~
wdt ) =
Xl (
iSl Ref
t) = - 1 - , so
J3 X2 () W2 ( t ) = - 21 Xl () t +2 t =
,/
Fi(.;. 12.7. 'J'ransients of controlled two-phase oscillator
W3(t) SOIJ t(' COI II 1'11 \'('.<1 tmnsientH of th(~ cont.rolled ()~iciJlalor n.re plott.c(l in Fig. I 'J. .'/, :, howilll': I.illl(· f'llldious :t:J (L), :(: 1. (t) iI.:, wdl aK t.\w pnt,h of (,\1(' st.aLc Wcl,OI ,1' (1) :"1 (1.) I .i:I;~ (/, ) i,1I Uw sl.aL\' pln,III' . TI II' .etII'V' ·S i,'I(I' icat,~ t.hilt. til(' (1 111:1 11 111.(11
lI 'il l"'I',
(12.18)
(Ir '!/I
= -21 xdt)
iS 2 Ref ---y;;,
(12.19)
J3 X2(t) = iS3 Ref -.
- -
2
Iso
A hlock diagnulI of the complete srw.(~d wn1.rol system is drawn in Fig. 12.8, \Vlu:rc Lh(: 1"( ' /'<0/'1'1)('(' siguais ror tllt' sLaLor ('IIIT( ~ ItLs nrt~ supplicd by the oscil hLor ; 1.1)(' 1I11I[1liLlltl(· is d( ~ riv(~d frolll LIt(· flllldioll 1',('II('raLor which iII tllJ'll is ('1/111.1',,[1,'" 11'y 1.11" d" !lir,'d 1'01.01' fl'('CjIWIIC' y I,) ~ "'1 r,J , I! ,Y IIInl\illl'. Lili' ,"t1.puL
250
12. Control of Induction Motor Drives
12.1 Control of Induction Motor
of the function generator dependent on stator frequency, field weakening can also be implemented with this scheme. The other actuating input of the controlled oscillator, YI = wlTo, deter mining the frequency of the stator currents is obtained by combining the measured speed signal W with the rotor frequency reference W2Ref as pro duced by the speed controller. To restrict the operation of the motor to the "linear" range, the rotor frequency reference is again limited as shown in Fig. 12.8. Thus the stator frequency is determined on the basis of the measured speed and the required torque. For larger motors having low pull-out slip (Sp < 0.1), the signal for the stator frequency is derived as the sum of two greatly differing quantities, WI = W + W2Ref ~ W, which makes the realisation of an accurate torque limit difficultj to improve this, the wI-signal may be formed digitally. ln fact, with the microelectronic components available today the complete control scheme can be realised on a digital basis. This is of particular interest with the field orientated control described in the next chapterj only a digital equivalent of the controllable oscillator is shown here.
according to Eq. (12.6), the angular position of the stator ampereturns- wave is correctly advanced under load, while the flux wave remains at its former position. This simple example shows how digital techniques can be employed to great advantage.
"
USI
I
I
1\1\r.~I\(d\r.Ofl"'III1AA~1I1I1I1
VlTv vvvvVv vvllVlI1JWVVIJ ~t Jml 1'\/)J\QAJ\I\OI\I\I\I\I\J\OOAO
C\J' V
cru \Tv V"'IV U4\n.nJ4JY9J v y~ t .,
USI
V0"'0/\ \fv"'Vr-ev, , 01\ 1\ 'L • I .m I 0Jl,J\JV>
1\VlJlJVv AAAA~i~Ll\l\oOI\ÔII"""" Vv v V1/ V1l1l 'lV'rJ ~ 'NQ1)
~M
......
CúRef
t
mL...
ir::-==.
~
~
/ Cú
:~
L_ .___ ~b
a
Controllaw
251
CúRef
---
M ---p
t
6001/min (1)2 Ref
Load feed-forward
(1)0
õ
sin/ cos Table
.,,,. "He' I I. , I I
(UI
(J)o
y
'SRef
ç
-~~1/min
Phase split "2 'S/Rei .~ ~ 7J---.
ISO 'S3Rel
ISO
(u
.
~0,5s--+l
300Aj O
! ;SI~
,I...... ~ ,~~~U'~~4wtmmm.u • ~.@~JW" 1
I
, , .. 1 "1"iI
"""
(Il O
."iK · 12.9. Digital signal processing for obtaining variable frequency, variable am plil.lld c
current references
Part of the digital control scheme is seen in Fig. 12.9, where sections uf trigonometric functions are stored as a function table in a memory chip, !.Iw address of which is the angle ( = J w1dt in suitable increments. The I.W() variables cos ( , sin ( moving on a unit circle are then multiplied by the OII!.PUI. of the function generator for the magnitude of the stator currents, I,crorc lhe signals are split into three phases. The dynamic response may be I';rca(.\y improved by a feed forward signal based on the load angle Ój this is :,illli\;n' t.o rq)lacillp; ali 1- by a Pl-controller. When choosing !.h(~
I)
Ilrc l. l llI 0h
'1'1/
(I ~ . :W)
c Fig. 12.10. Transients of induction motor drives (a), (b) Simulated results for starting, loading and speed reversal with the control scheme in Fig. 12.8. (c) Measured transient of experimental drive with the digital speed control according to Fig. 12.9 The control scheme shown in Fig. 12.8 was simulated OTI a digital computer with Eqs. (10.50)- (10.53) representing the motor, while the controlloops for t,he stator currents were simplified by unity gain, first order lag elements ['1'111. 'Th(~ \"(~S1l1ts are plottcd in Fig. 12.10, showing a starting transient with sUI>S('qlH ~ lIt ]oadillf!; anel a. s!)('('d f( ~ v(, .. :"d wheu inverting the speed reference. f)1I(~ t.o til(' rol.o!" rl'( ~ q'\( ~ J}( : y lilllil., 1,111' 1i 1H""arly wil.11 I.illll' " II. iII illl.,·rc· til ,i lll ~ !.o JlO!." Ll lld. 1.I1l' JlI:II',lld.i:; iJlI', cllrJ'(~Jd, iii rairly
2;' 2
well maintained during the transients, even though the design of the control scheme is based on a steady state mo deI of the motor. A digitally controlled experimental drive using a 5 kW induction motor sllpplied by a voltage source transistor converter with fast current control I'MS tested with the control scheme in Fig 12.9; a recorded speed transient with a high inertia load is shown in Fig 12.10 c [H20]. Clearly, as the stator lú:quency is tracking the speed, the motor cannot be "pulled out of step" IIII t, when overloaded, operates at the torque limit. Even though acceptable results can be obtained with a well tuned control L11at is based on a steady state modeI of the machine, these restrictions are 11Ildesirable for high dynamic performance drives, they may be avoided with I.lw control strategies described in the fol!owing sections. Basing the control () Il the dynamical models of the machine, AC drives can be designed whose dynamic performance is equal or superior to that of DC drives, considering Lhe larger bandwidth of the converter and the reduced inertia of the motor.
be measured with cage motors, it is appropriate to substitute for the rotor current vector iRe jE an equivalent quantity that could be measured with stator based sensing equipment. A good choice, as will be shown later, is the rotor fiux, Eq. (10.32), or the equivalent magnetising current vector, defined in stator coordinates,
'P.R(t) e j E = LoimR(t) = Lo imR(t) ej
12.2 Rotor Flux Orientated Control of Current-fed Induction Motor
mM(t)
W: I ~:
3 Lo 1m [:is (i.R eJ E)*]
ol,I.; l.illl'd hy
1.111' ti 11 X
W:I.V' ·
(12.21)
d('~j('rihillr; tlL
1" :: IIILilll',
rrOll1
i'i LaLor
CIIIT'·III.:i .
III'(,W( '('II
Lilc roLor · CllITCIILH alld
SI IH '" 1.II" · rol.or
.
= ~ ~ 1m [is (imR 3 1 + O'R
is)*],
(12.23)
~ (1 -
(12.24)
0') Ls 1m [is CnR] .
1nserting the magnetising current vector according to Eq. (12.22) yields
'I'11(! heuristic control scheme described in the preceding section is, in spite of L!te acceptable results obtained, not best suited for high dynamic performance cI ri ves . Because of the open loop flux control it would be difficult or impos ~: ii>I( ! Lo operate the motor with ful! torque at low speed or even standstill w lli (' il (!xcludes, for example, its application as a servo-drive. o remove these I (·::l. ricLions it is necessary to return to the dynamic model equations of the I 1Ic1Ilcl.iOIl motor, based on instantaneous currents and voltages, as derived in ~ ;( '(1.. lO .l. Of particular interest must be the interaction of stator and rotor "lIr)(!II L~ resulting in fiux and torque. When initially assuming that the stator ('11 rJ'(!n Ls are impressed by fast current control loops considerable simplifica l.iollS result since the stator volt age equation (10.50) becomes a concern of 1.11(' cnrrent controllers and can be omitted from the model equations of the cI ri ve. The restrictions of steady state are now avoided which is a precondi I.ioll for achieving rapid transients. The power supply could be a transistor collverter with high switching frequency and adequate ceiling voltages or a cyc!oconverter, resulting in very fast response of the stator currents. The expression for the electric torque (10.48) =
+ (1 + O'R)iR ej E]
which is simplified to
12.2.1 Principle of Field Orientation
'/Il M
= Lo [is(t)
Eliminating the rotor current from Eq. (12.21) results in
mM(t) =
.
Ii(t)
(12.22)
mM(t) =
2
253
12.2 Control of Current-fed Induction Motor
12. Control of Induction Motor Drives
CIII"I"I'1I1.:i (';1.1111111.
2
3 (1 -
.
0') Ls imR 1m [is e- J li] ,
(12.25)
where
is e-j li = is e j
Ó
= iSd
(12.26)
+ j iS q
rcpresents the stator current vector in a moving frame of reference defined \)y the rotor flux or the magnetising current vector imR · The angular relationships of the current vectors are depicted in Fig. 12.11; Lhe stator current vector in field coordinates or the "field orientated" stator wrrent vector is seen to consist of two orthogonal components,
iSd = Re [is e-j li] = is cos Ó iS q = 1m [is e-j li] = is sin ó in the direction of the magnetising current vector and the other perpen dicular to it. ln steady state condition iSd and iS q are constant apart from ('(IilVerter induced ripple, i.e. the stator current vector is and the magnetising (,IIIT(·m t vector imR rotate in synchronism. The steady state current trajecto til!S iII Fig. 12.11 were taken from actual measurements with the drive shown
III1C
iII I"ig. 12.28. IIlt,rod\l cing ;t
the quadrature current component into Eq. (12.25) leads to si III pie (!xprcss ioll ')
'III'A/(/.)
/; i'"II ·i.,';,/,
f..:
~ (I
:\
0)/' ::,
(12.27)
:!51
12. ContraI of Induction Motor Drives 12.2 Control of Current-fed Induction Motor
di T R -mR dt
" Measured steady state
Current traj~onés
I
(12.30)
dt = WmR(t) ,
d( dt
d8
= Wl (t) = WmR + dt
Multiplying Eq. (12.29) by sults in dimR TR ~
UI! iII
+ Lo dt [P + O'R) iR + is e- J ~J v
i --mR
wiL/) '1'11
e-
J
1' 11/ /lu , tlli:-l J(~ads to
c
e-jl!
and expanding the left hand expression re
' + J. WmR TR '2mR + ( 1 - J' WT) R 2mR = '!.S. e - j I! ,
(12.32)
which may be split into real and imaginary parts,
which gives a hint, why the transformation of the stator current vector into fídd coordinates, also called field orientation, is the key to rapid control 01' AC machines. This principIe has been proposed by Blaschke [B33-B38J, ( ~ xtcnding and generalising earlier work by Hasse [H26J. Clearly Eq. (12.27) reminds one of the expression for the electric torque of d DC machine, Eq. (5.3), where i mR corresponds to the main flux
'1'0 outain the complete model of the AC motor in field coordinates, i is
mR
IHIW IIscd for elimination of the rotor current from the rotor voltage equation (IO.rd) witlI UR = O, ,
(12.31 )
.
Stator axis
Fig. 12.11. Angular relations of current vectors in steady state
d
(12.29)
d{!
/ Rotor axis
isa
.
= Is ,
where dé/dt = w. The definitions (10.7) and (12.22) yield also the instantaneous angular velocities of the magnetising and stator current vectors
~/\/ ~s
. + (1 - j WTR ) '!.mR
255
= O;
dimR. . T R - - + 2mR = 2Sd , dt d{!
-d
t
2S q
= wmR = W + T 2.
R mR
(12.33)
= W + w2 .
(12.34)
Eqs. (12.27, 12.33, 12.34), together with the mechanical equations (10.52, 10.53) constitute a model of the induction motor in field coordinates, as Jescribed by the block diagram in Fig. 12.12. The two field-orientated input currents are produced by transforming the stator currents into a coordinate system defined by the flux angle (!. This is performed in two steps, first converting the three stator currents to an orthogonal two-phase AC system,
Ú(t) =
iS l
+ iS2 eh + iS3 ej 2'Y
= iSa
+ j iSb .
(12.35)
With the condition for balanced three phase currents, t.llÍs results in
iS l
+ iS 2 + iS3
= O,
iSa(t) = ~ isdt) , iSb(t) = ,;; [is 2(t) - is 3(t)]. '('!tese AC currents are then, by transformation into field coordinates, con vcrtcd to DC quantities in steady state
(12.28)
1:.,, (t) h(·IJ(".( ~
e-,i
(!
= (isa
+ j iSb) (cos (! -
wc !tavc 'i Sd(l.)
'I.":,, cus U 1- 'i s /. silJ
";-" IU)
i:.'" ( ' 0 11 f'
'1' ,' :/1
e,
f!iul'.
j sin{!) = iSd
+ j i Sq ;
(12.36)
2S6
12.2 Control of Current-fed Induction Motor
12. Control of Induction Motor Drives
COSp
sin p -----------l
I.
IU!!!...JI
sq '
of control in field coordinates as suggested by Blaschke, is illustrated in sim plified form in Fig. 12.13. By modulating the signals at the control side with the reconstructed ftux angle (e jll ) and subsequent 2/3 phase splít the corre sponding interactions in the control plant are cancelled as long as the ftux angle is correct and the delay of the converter and the residual lag of the current loops are negligible, i.e. the converter acts as a current source. This part ignored for control design
1"-
Field coordinates
S/ator coordinates Control -+I+- Converter --+t--Motor
.1.
References
is R6/ 3
Stator coordinates
Coordinate transforma tion
257
-I
'1'
Field coordina/es
J I
Field
coordinates
Fig. 12.12. Block diagram of induction motor in field coordinates, ,\ssuming impressed stator currents
The fiux angle {2, obtained by integration of Eq. (12.34), is an internal variable; the coordinate transformation essentially constitutes a demodula Lioll. Once the transformation has been completed, the dynarnic structure of Lhe AC motor is quite straight-forward, not unlike that of a DC machine. ln pa rti cular there is a lag time constant T R comparable to the field lag of a DC 111<1,chine; therefore the magnitude of the rotor flux or the equivalent mag IIdisill g; current vector, imR, is not suitable for fast control action infiuencíng LO!(Jll<' , this task should be assigned to the quadrature current iSq. The leak ",'," I.illl'~ constant in the quadrature axis, which would be the equivalent of 1.11<' 1.(; ar mature time constant, is not effective in Fig. 12.12 considering the i l.:::: llIlWd irnpressed stator currents. I'()l' achieving high dynamic performance control of the AC drive, the sarne :·:I.ratc'g;y as with the DC machine is appropriate: The magnetising current ,i II should be maintaíned at maximum levei, límited below base speed by sal.llratioJl of the iron core and above base speed by the ceílíng volt age of 1.I1(~ converter, while the torque should be controlled through the quadrature CllITmt. When highest efficiency of a drive is of utmost importance, as in I,al.l ,c ~ r y snpplied vehicles, the fiux levei could be reduced when operating the IIlOLor at part load for extended periods, in effect balancing reduced iron Iw;,~cs against increased conduction losses in the motor and converter. This was l\f( ~ lL t i oll Cd in Sect. 5.3.3 on combined armature and field control of DC lll,d,o!'s. Of co urse, there is an increased c1clay wIteu the flux levei first needs to 1)(' raised iII case of sllddc-~ Jl torqm~ delll illlll. 1,'01' all,'viaLilll ~ 1.1[(' COlllpkxil.il's or 1.1](' I'I C ('(lIII,rol plnlll., iI. call lIOW Iw al:
Feedback sianals
Fig. 12.13. Principie of control in field coordinates
Microcomputer ~ Converter and motor
-·-·------- ·- '- --·- ·-·-·-·-·- -- '-I ·- ·1:1 I, . :iSb
r
Flux model Converter with current control
III
1."1111>1,(''' 1.0 c;lllI·d 1,111' coordillal.,' I.rilll :1 f c!lllv,'rl.l'l' j 1.\1111 i d pl ~) whjl(' h Iii "fl.I'1.
Field weakening
Flux control/er Control
Field coordinates
.:.
Statorcoordinates
Fig_ 12.14. Speed/ position control in field coordinates of AC motor
Wit.h 1.l!C'hC' C:OllW ~
a~sullIJltions
I" i 1'. ' I '2 . 1'I
r" r
II.
the design of the remaining control system be
1.0 I.hat. of a DC lIlotor coul.rol schcme; this is shown in 1''': " Li. III .. " lIt]',,1 cOIlI-iI':1I 1'11.1. io II . '1'1 1\: /. h !'PI ~ 1:11 rrt~lll, r cr(~n: ll c (' s ror
\ ' ( ' I' .v :; illIiL u '
258
12. Control of Induction Motor Drives
12.2 Control of Current-fed Induction Motor
the controlled converter are obtained from the two-phase current references {isa, iSb)Rel with the help of Eq. (12.35), ·
2S1Ref ·
2S2 Ref ·
tS3Ref
2 . = 32SaRef 1 .
= - 3" 2Sa Ref
+
1 .
= -3
2Sa Ref -
1
.
1
.
v'3 2Sb Ref v'3
(12.37)
,
2SbRef·
I)r<~ ced ing this phase splitting operation is the modulation of the signals calling for four multiplications with trigonometric functions of i.lw lInbounded flux angle e' mod 27r, (1,,;,1, iS q ) Rei,
(-i ';a
+j
iSb)Ref
= (iSd
+j
= iSdRef COS e' -
iSq)Ref iSqRef
ej~'
sin e' + j
[iS qRef
COS e' +
iSdRef
sin e']. (12.38)
'I'ü adüeve cancellation of the machine-internal demodulation, the available
Ihl x signal e' must be in dose agreement with the actual position e of the fllndamental flux wave; this aspect of flux acquisition still has to be discussed. Proceeding in Fig. 12.14 further to the left, it is seen that the remaining colltrol structure is the sarne as that of a DC machine. lnstead of the armature Clll'rcut controller, a torque controller, whose synthetic feedback signal is deli vcd on the basis of Eq. (12.27) from flux and stator currents, generates /.Iw (llIadrature axis current reference. Speed and position controlloops may I II' s llp( ~ rimposed. With a digital realisation, the position controller can be a Nilllpk proportíonal controller, as will be discussed in Sec. 15.2. It would also 1)(' po:,sihk Lo omit the torque controller and generate iSqRel directly with UI(' :'11\'1'.1 ('ouLroller. III LIli ~ .I i rect axis there is a flux controller, the reference signal of w hich is 1,'.111< "' .1 :tl)()v(~ base speed to achieve field weakening; the flux reference also ::,'IVI':-: Lo n~dllc(-~ the torque limit at high speed. WII!'JI r<~calling the mechanical IDodel in Fig. 12.1 b, the strategy of field olw(JL;tLiou is as follows: When the speed controller determines that a certain V:dlll' 01' Lorque is required for maintaining the commanded speed, references rol' (, II(~ Il(~ cessary tangential and radial forces are computed which are then, hy coordillate transformation, converted to reference quantities for the forces fi, h, h· The transformation must be based on the instantaneous POSitiOll 01' UI(: piu ; hence the need for flux acquisition in a field orientated control r:dICIJlC,
CII'ar\y, the principIe of control in field coordinates looks like an effec Li V(' way of decoupling the complex multivariable control structure of the i 11.1 II cl.ioll rnachine. Its main advantages are:
• Di n~ct access to flux and torqu e pennitting controlled fidd weakening anel Lor
:1l'1I1' ,·"",Lrult.-d fUl' \ ("IUlI IO!. h~ ~
I'H II""I
ullL lIr ll I.o-p ;
1.I1 1Jl
( '()l'I'(' ~l p(lll<\ fl LII
I.Iw
259
function of the commutator on a DC-machine with current limito When overloaded, the motor is stalled with maximum torque. • ln contrast to the schemes discussed in Sect. 12.1 the decoupling is effective in dynamic as well as steady state conditions. • ln steady state the drive controllers process DC quantities which makes the controlless sensitive to unavoidable phase shifts. The speed control system contained in Fig. 12.14 has been simulated on a computer, using the sarne set of parameters as for Fig. 12.10a,b [T14] . The current control loops were again approximated by unity gain first order lag elements. When comparing the starting, loading and reversing transients in Fig. 12.15 with those of Fig. 12.10 a,b, the superior performance of the field orientated control scheme is obvious. The flux is maintained constant and the electrical torque shows a rapid rise, limited only by the lag of the converter; when the speed controller is damped, the torque is precisely fixed at its maximum value. USl
/51
~c>c,J'MbJV\NJVV\ ~ ~ ~ ~ fi fi/ i vlJYvvW • I
o
~
/51
C>
.
t
.
voWVVt "
VífJvAVbJVI/VVV\}V~VÔ\[\NVV ~~~~ •
II A
t
mM
~
b
a
t
/151
f\. /\ Wv VlJ'CTV~fI~AoVJV. I
~~-----------
\TvVVVllVV ..
,fifi
t
())
v# •
\1\111\ fi f"\ VQlJ\/\JCJf"\ fi fi II fi fI/\ fi
fi
(I/mM
1\ fi fi ~ Ofi VVVl)lJ
Ilm Rl
t
mRl
f\f\flflfIfI fl /\ /\ 1\ fi fi VV lTvvvJVV\N'v
fi
JV1fJ'vo " " fI/
=-:J
~ ..t
Fig. 12.15. Computed transients of AC motor controlled in field coordinates
Of course, these are only simulated results which have been computed w ith a highly idealised model of the control plant; to realise a field orientated
colltrol scheme in practice, several serious problems first have to be solved: • Acquisition of frequency independent flux signals, representing amplitude and position of the fundamental flux wave. • I ~(fects of residuallag of the current controlled converter. • Iluplelllentat.ion of the complex signal processing shown in Fig. 12.14 . • 1';stillIatillf'; the cffeets of a detuned flux model. 'I'\\I'iil" probklll:'l will
1I0W h( ~ di:'WllS~;( ~ 1l H(~ p;l.r1l.t.( ' \y.
260
12. Control of Induction Motor Drives
12.2 Contro! of Current-fed Induction Motor
261
12.2.2 Acquisition of Flux Signals Clearly, having up-to-date information on the magnitude and phase of the fundamental fl.ux wave is of paramount importance since this is the basis or coordinate transformation, leading to decoupled control of the currents 1 13 37,L7]. For obtaining a frequency-independent fl.ux measurement one could at t.(:mpt to measure the fl.ux density in the airgap of the machine directly by plaocing suitably spaced magneto-galvanic or magneto-resistive devices such ;lS I{all- sensors on the face of stator teethj by interpolating the local samples 01' llux density, an estimate of the magnitude and position of the airgap fl.ux IVave could be obtained. Then by adding a volt age component proportional t.() Lhe stator current vector, a signal representing the rotor flux '!f ej f or the R 11l 0.5 Hz), where the drift of the integrators '·V" IIL lliI.I\y lJccame too largej this limitation prec1udes the application of this ::..!t(:lII(' to dri ves requiring position controI. Again, a special1y prepared motor w",ild llave (;0 be used. ' I'h,' las!; Illcntioned disadvantage is avoided by measuring terminal volt :'1'," :;, io':o IlSillf!: tile stator windings as sensing coi Is. This, however, cornplicatcH 1.1 11' ,; illl:l.Lioll evell [llr1.b er l)(!CélllS(: of tlw n:sistive voltage dl'op, tilat dllll1" i,,;!.I.":; aI. low J'n '<JIl(\lLcy , lll(\ llIU s l. hc (·olllJ)(:II:;;t1.(·d prior to illt.q( ratioll j a.'i· Lili ' !il.; do",. I'<',;i :;1.; I.LII·(' "halll!;<'S wiLh l<'nq li' 1'11 1.1\1'1' Ih iN IIH' I\IHll'il;l ~ HrlH' IlH: ClI,O o 1"" '1111 11' ",lIiLl ill v,.J v,·d . '1'1'1'1 1.:1 IV tl.b,,"1. ~o l ·I III" 'l( l l.\lrl ' ( ' '''II~I(·I HJ ~I.i ,," IIII.\fI· jlld i
iSb
isa
imRa
a
b
Fig. 12.16. Flux mode! for obtaining the magnetising current vector (a) in stator coordinates and (b) in fie!d coordinates
cated a practical lower frequency limit of about 3 Hz with a 50 Hz motor. The situation may be improved when performing the integration in field co ordinates, as discussed in 8ect. 12.5. Another quite different approach is based on the model equations of the motor. The equation representing rotor voltages, Eq.(12.29), a first order vectorial differential equation, may be used for computing 1mR(t) in stator coordinates on- line. The measured stator currents is and speed w could serve as input functions, with the motor being described by a single parameter, the rotor time constant TR. By again introducing orthogonal two phase current components
ls(t) = is(t) ej «t) = isa(t) + j iSb(t) , 1m R(t) = imR(t) e j e(t) = imRa(t) + j imRb(t) ,
(12.39) (12.40)
the block diagram of Fig. 12.16 a results which illustrates in graphical form the interactions of Eq. (12.29). The structure is similar to the controllable oscillator shown in Fig. 12.6 for Yl = wTR , Y2 =- 1. A more direct method of computing the magnitude and position of the fundamental flux wave is to immediately convert the flux model into field coordinates, as described by Eqs. (12.33, 12.34) [841,842). Again, stator cur rents and speed serve as input signals while the transformed stator currents is d , iS q and the magnetising current i mR are now directly available as output signals for controlling fl.ux and torque. The fl.ux model in field coordinates, Fig. 12.16 b, is preferable because of its simplicity and potential accuracyj also the output signals are constant in steady state. Of course, the infl.uence of the rotor time constant T R is still presentj this parameter may have to be Ilpdated during operation. Fig. 12 .16 b corresponds directly to the model of lol,,! IllOtor showll in Fig. 12.12 and is inc1uded as a "flux model" in Fig. 12.14. TIl<' IIS( :
262
• Flux signals are not based on local measurements and are not affected by slot harmonics and stator resistances , • Use of a standard motor without additional sensors is possible. • Flux sensing is operative down to zero frequency, because no open ended integration is needed which would be subject to dríft. • The model immediately generates the variables needed for feedback control.
For the following it is assumed that the first approach has been taken but that its potential is exhausted, leaving a residuallag Te that cannot be reduced any furtherj hence the second option must be pursued. The situation is depicted in Fig. 12.17 showing the two opposite transfor mations which are supposed to cancel and the unavoidable lag effects placed in between. With the vectorial notation for the currents in stator- and field
Ilowever, there are stringent accuracy requirements, for example with respect Lh e modulo 27r integration of lIj this can best be solved by digital com )llltation in a microprocessor as will be shown later. The dependence of the roLor model on the time constant TR constitutes a source of error because ;\Il illcorrect orientation li' of the flux wave leads to undesirable coupling be Iwcen the d- and q-axes and could endanger the control in field coordinates, CV(~ lltually even resulting in instability. Another possibility for obtaining a flux signal is the evaluation of the sta Lor voltage equation, Eq, (10.50), when sensing the terminal voltages and cur II IIIL5, However, after eliminating the unaccessible rotor currents, Eq. (12.45) n ls lllts which contains the rotor flux in derivative form so that an open ended illLegration would again be necessaryj because of the necessary isRs- com IWHsation and integrator drift this is likely to produce unreliable results at low speed. This option is not pursued here, it will be further discussed in ~;(ICL. 12.5. L()
1 2. 2.:~
W 11\'11
263
12.2 Control of Current-fed Induction Motor
12. Control of Induction Motor Drives
Field coordinates
Stator coordinates
I i'SqRef I I iSbRef
I
iSqRef
+
from controlfer
iSaRef
F·, LSRef
F· LSRef
i sq
I e'jp
i'SdRef
Decoupling
Field coordinates p
iSb
e jp
1+j w l Te
iSdRef
~
+
~
JSRef
+
F·
Js
+
Converter with current control
Drive contrai
iSd
isa
!s
Motor
Fig. 12.17. Removal of undesired coupling effects
caused by the residual lag of current control loops
l<:; lfects of Residual Lag of the Current Control Loops ex plaining
the principie of control in field coordinates, Fig. 12.13, it ;,.ss llllled that the stator windings are supplied from perfect current ::0111'('( ':: , i.(I. L1tat the lag ofthe current controlloops can be ignored. This calls 1'11 1\ (''v''locollverter or a volt age source converter with high frequency pulse \Vi d I.h 1111 )(llllation, allowing rapid access to the currents through the reference 1I 1!';lIil.l~;, as well as adequate ceiling volt age throughout the operating range. (:I(·;l.Ily Lhe second condition is in conflict with an economical design of the ('(111 wr!.cr uecause it would preclude a full utilisation of the available voltage III higll speed. On the other hand, if the lag of the current controlled converter ii: 1I( ,i, IIq:~ligible, the decoupling by the inverse model becomes inaccurate and 111)(1('siraiJle coupling terms may arise which will eventually render the field ol'J('ll!.al ,(~d contro! inoperative; to avoid this, two options are available: 1111:: 1)(,(' 1I
coordinates
is(t) F is(t)
+1
S T 2 .9 -/-
iii 1.11(' 1:('/'1'1'1.'11('.(:
1'
ch;\IIJid
'1'2 < TI,
.
lSRef
01'
t.Iw sLa!.of
('llIT(~H(,
('.(lJl!.J.'o!!oops.
• t\ddil.i'llI (II' d('I'lllIpfillf:' (.('1'111:: iII lidd (,ool'dill :iI,I '': , i.I'. prior (,0 t.lH~ l.rau I: r'Il 'llI liI,i,'1i1 1111',0 ill,nl ... r (·o"ldlllal." Ii.
(!
isq(t)
=
F ·'
LSRef
j
e
(12.42)
(!
lhe lag effect of the converter may be described by Te
(12.41)
= is e- j = iSd(t) + j
iSb(t)
and, correspondingly, for the reference signals,
• IlIdll sioll of a lead- lag network with the transfer function
I" (.~) -= '1'1
= is(t) eH = iSa(t) + j
dis
.
.
dI + 15 = 15 Ref .
(12.43)
WiLh the assumption that the lag time constant Te has already been reduced a relatively small vai ue, the undesirable coupling is mainly due to phase Hh ir!. ra.t.her than amplitude. ln steady state condition, with is = const., (,!",/( ('-'I and inserting Eqs. (12.42, 12.43), this yields Lo
"' , '1,,'\
,.. ./ I 1
.i I.! I 'I:.
1. -" Ikf
( 12.11)
264
12. Control of Induction Motor Drives
12,2 Control of Current-fed Induction Motor
when specifying a decoupled transmission, F·
I F · !SRef'
'!.s ~
Lhe result is F Ls 'I Ref = ( 1 + J' Wl T)F. e !S Ref
.
,(,his means that corrective terms should be applied to the output signals of I.he controllers. Written in orthogonal components, this calls for LSdRef
= !SdRef -
Ls q Ref =
is q Ref
Wl
+ Wl
+ u Ls
ilis
dt
+ (1- u) Ls
ili mR
dt .
(12.45)
111 :: I.c' ady s t.ate condition, i.e. with constant speed and sinusoidal currents, 1,1,,"1
= Wl,
is
= const ,
imR
= const.,
11('11 C'C' I,,: ; (t )
=
(Rs + j
Wl U
Ls}is + j
Wl
(1- u) Ls i mR
.
(12.46)
'I'his (:()lIflrms that the fundamental component of the stator volt age rises with rr('<jl\c~nc:y, Fig. 12.3 b. When the available voltage range is near1y exhausted :d, the frequency WlO - some margin must be left for control - a further illc['( ~as c oE" the speed is only possible by reducing the magnetising current, ic', iJy fidd weakening, This can be achieved with loop scheme employing a sp eed dependent reference imRRer(w) a~ Fi~ . 12.1401' • :1.1\ ô\.lI xil iary ("()lJlrollooJl whidl lilllits Lhe 1J1 I\f';J lil.1Jde of the slalor v()ll.agc~ hy J'( ' cllI<:iIl,~ I.II( ~ flme r c fer ClJc:c 'i ",.{l/i"J ; il s ilJlíl llr Iidwll\(: h: \,n c : arlí( ~ I ' 1)('1"'11 clC'He'I'i1lC'd for I){: 1I1:u,ltinc'H, I,'i" , 7. 1:\ • :1.11 OpClJ
tdlOWIJ iII
When the field is reduced, the ceiling voltage of the converter should nor mally be fully utilised, resulting in gradual cessation of pulse-width modula tion so that each leg of the converter eventually produces square wave output volt ages, Fig. 11.6. The stator currents would then be no longer sinusoidal but the field orientated control still remains functioning as the computation of the flux and the magnetising current is based on the measured stator currents, Because of the limited voltage and stator currents, the maximum torque is, of course, reduced in the field weakening region, Eq, (10,85) . 12.2.4 Digital Signal Processing
Te iSqRef, Te is d Ref .
The practical realisation of this additive correction is quite easy since DC sig Ilals are involved in steady state. For simplification, the speed signal W may Iw used instead of Wl with low slip motors. ln view of the limited ceiling volt age of the converter, the residuallag Te of the current controlloops is likely to increase with the stator frequency but if the control is implemented in a IlIir.rocomputer, this effect can be offset by varying the decoupling parameter 'I;. as the speed changes. The rating of the converter is determined by the specified values of stator voltages, currents and frequency. After eliminating the rotor current from I';qs . (10.50, 12.22), the stator volt age reads . II.,>' (/.) = Rs!s
265
The signal processing required for the field-orientated control, Fig, 12,14, as well as the flux acquisition using a dynamic model, Fig. 12.16 b, is of consider able complexity; in particular, the various multipliers and function generators needed for the coordinate transformation are difficult to adjust accurately when realised with analogue components. On the other hand, on-line digital control was economically impractical as long as it had to be performed by large and costly process computers. These obstacles have delayed widespread application of the sophisticated control schemes needed for AC drives [L44]. However, this situation has changed dramatically since digital processors have now shrunk to minute size thanks to the progress of microelectronics. The reduction in volume and cost and the advances in processing power have in the meanwhile reached a point where ali the signal processing required for a high dynamic performance AC drive can be executed by a single micro processor which, together with the associated peripheral components, finds room on a postcard sized printed circuit board. The main advantage of digi tal control with microprocessors is that the sarne standard hardware can be llsed for many different applications because the function of the processor is determined by a flexible and individually adaptable program, i.e. by soft ware; once the program has been developed and verified it can be duplicated at negligible cost and transferred to a read-only memory chip connected to Lhe processor. A precondition for the digitalisation of signal processing is, of course, an adequate resolution of the signals with respect to amplitude and time, Le. s ufficient word length and computing speed. The current 16 bit-processors are qllite satisfactory from these points of view; a 16-bit representation of vari ; ~blcs corresponds to an internal resolution of one part in more than 65,000; ir 10- or 12- bit analogue-to-digital converters are employed for converting I.IJ( ~ analogue signals from the plant, such as stator currents or motor speed, I.hc: d fec ts of quantisation are usual1y unnoticeable even though there are s l.il! exc c ~ pti o ns ) when higher internal resolution, such as 32 bit, proves to be llC'n'ssary wi th some variables. Most microprocessors still use integer valued IIl' il.hl\\(:(".l.i(" Inll, proo ,ssors witlt Jloatinr~ jJoint arit-hmctic are becoming more {,
266
12.2 Control of Current-fed Induction Motor
12. Control of Induction Motor Drives
levellanguage such as in C, to make them independent of the processor hard wa.re used. Reduced word length, such as 10 bit, is usualIy adequate for the I) I A-converters, i.e. at the output of the microprocessor because the DI A converter is part of the plant and thus under the surveillance of the digital controlIer, whereas errors in the AI D-converter produce false signals. The choice of the required resolution in time is more complex as it depends 1101". only on the type of drive but also on the particular function within the drive controi. Clearly, when sampling and processing non-sinusoidal currents h;wing a fundamental frequency of 100 Hz, a higher sampling frequency is 11("( ~d ed than for the digital realisation of speed or position control, where the ral . (~ of change is restricted by the inertia of the drive. As long as computa I.ioual speed of the processors was a major consideration which is no longer I.h( : case with todays microprocessors, an advantage could be gained by em ploying different sampling periods for the different tasks. For example, for an ('-"qJcrimental 1.5 kW induction motor servo drive having a PWM-transistor power supply and an 8086 control unit, the folIowing sampling periods proved cOlllmensurable with fast control transients [841]. a) Sampling period of 1 ms for AI D-conversion of stator currents U pdating of fiux model COllversion to field-coordinates Cmrcnt and torque control Couv(~rsion of di q-currents to stator coordinates Olll.pUt. of current references 1» ):;aIJlplin~ period of 5 ms for 1,'1,1 x c01lt.rol including field weakening ,', pc 'c ,Ii control I'm:il.ioll control fi :;il.1I1 pi ill['; time of 1 ms for the stator currents is adequate up to a stator r""I[IIC :llcy uf about 100 Hz, when the currents are sampled and processed ten l . illlC'~: per periodj at this frequency the fiux vector is computed at 36° inter v:th wllidl can make the use of a simple extrapolation scheme desirable. At :: till Ili g h(~ r stator frequency, faster sampling of the variables is required. For I.odays hi,..;h dynamic performance servo drives, a sampling period of 100JLs, ('oIT('spollding to a 10 kHz sampling frequency may be specifiedj this can be adlil:v(~d with signal processors. Another possibility are application specific illl.l'l~ra.I,(~d circuit (A8IC's), where ali the control software is incorporated as I'IISt.Olll- ltardware on special chips, [K34 - K38], [TU]. A II illcrelIlenta.1 position sensor as described in Chapo 15.2 can serve for :i<'lI::jllf~ iLJlgular position as wdl as speed; wlwn tlw [orward/reversc pulses are II.I ' c '. 1I1111lIat.( ~ d iII a r<~vcrsihlc cOl.luter, which is illiLially set hy a. rd(:n~llce pulse, LIli' :!.III-,lIla!' pw:itioll o[ lhe shaft call 1.1\' cld,('d,c'd ,II. 1.11<' salllpliJl UillSl.il.lll.S hy 1.111' 1'11 ic 'r"l H'IlC' \ ':-lHOI', [<'ol'll1il)(( !.Ii<' ditl" ' I'\'III ' " II('!.w('I'1I I.WI) Hl rll Hl'lflWUI. Nnlll(d('N 1" ""icl,';! II II 14'11111lI'c' "r I,!t., II V('l lIf,\" :1 1'1"1'01 i II 1.1'1' 1111' 1. illl.\'l'vnl. ~lI.l.lIndl'y, fI,1I
267
encoder, delivering at any instant the absolute angular position of the shaft would be preferable for speed and position sensing as it removes the ambiguity with regard to initial position, but this advantage has to be balanced against increased cost and complexity. With a high resolution incremental speed/position sensor producing 8000 increments per revolution, the speed signal has a frequency of 200 kHz at the nominal speed of 1500 1/min for a 4-pole, 50 Hz motor, Hence a 5 ms sampling period of the speed controlIer results in a speed measurement of 1000 counts at nominal speed but correspondingly less at lower speed, 8ince this is usualIy inadequate with regard to resolution and accuracy, a quasi analogue sensing scheme providing finer resolution at low speed could be employed [845, KU]j another option is to convert the sensing scheme from frequency- to time-measurements which alIows better resolution at low speedj this is discussed in 8ect, 15,2. The design of linear sampled data control systems is part of the general control theory that has been exhaustively covered in the literaturej there is no lack of proven design methods, e.g, [43], A look at Figs. 12,13 and 12,14 reveals that even though the control plant is highly nonlinear, a considerable degree of decoupling and linearisation is achieved through the method of field orientation. ln addition, the principIe of cascade control, identical to that applied with DC drives, serves for consolidation by allowing a step-by step design of the control structure. All that has been said about the design of controlled DC drives, including the combination of feed-forward and feed back control, remains applicable here, because the difference between a DC and an AC drive is, besides the changed parameters, confined to the block "Torque controlloop". Further simplification is due to the fact that relatively high sampling frequencies can be chosen with a suitable microprocessorj as a result, the control functions in the outer loops are practically continuous so that the added complexity of sampled data design procedures is avoided. As an example, the difference equation of a PID-controller relating the sampled control error e(v) to the actuating signal y(v) may be written in straight parallel form
y(v)=G
TV T;[e(v)-e(v-1)], e(v)+Ti~e(JL)+
[
1
(12.47)
where T is the sampling interval, G the controlIer gain and Ti, Td the time collstants of the integrating and derivative channels respectively. The equa I.ioll is to be solved in real time by the microcomputer. The integrating term would normally be computed recursively in the form y, (II)
dlt) = :'1 i ('/ - 1)\ 1'(1/) .
(12.48)
'268
12. Control of Induction Motor Drives
12.2 Control of Current-fed Induction Motor
269
lt is important that all the variables are bounded to prevent numerical over
flow with integer arithmetic, or undesirable signal overshoot if floating point arithmetic is employed. The discrete transfer function of this controller is, letting z = eTs ,
_ Y(z) _ E(z)
F (z ) - - - -
G[1 +TiT-z-1 -z - +TT- -z -1] d Z -
1 - 500 ms -----I
!l ~ ~
1j
vrvv IL
(12.49)
\2 .2 .5 Experimental Results
3500 l/min
1"i,;llre~
12.18 and 12.19 show speed and position control transients of a 1.5 k W induction motor with a switched transistor power supply and micropro ('(~ssor control; the motor was not loaded during the tests. The 8086-based cOlll.rul algorithm comprised altogether 8 k bytes of read-only-memory [S44]. 'I'lw response of the drive is illustrated by the fact that after a fraction of (llle revolution the motor reaches field weakening speed. The smooth and rapid transients, resembling the simulated results in Fig. 12.15, indicate that tid
;eVOlutions
i ro
\
1~
b
~
V
iSq
•
~
mR
Field weakening./
i
~
20ms }-o- - -'"
frL
liRM'/' ,., ln "i ".wHil\4~'IH li , sq
i
(ú
10001lmin
/_/
q
.
f
....
·1000 1Imin
.
1"lfIM••
,:rt ..
I l U . ". ...
AIlU\~11I \
== iS2
~
Fig. 12.19. Step-response of a position controlled 1.5 kW induction motor drive
is,,,
=d
C\ V /\. V~~
~ . A --
12.2.6 Effects of a Detuned Flux Model
is,
V L\V A \ JL"">'J/"""'.\.J
10" 11-<. I'... . IH . N,) IO~ L" >1 1"'('.1 f('v('r~al "I" 1.:' I', W illdlldiollllllll.or wl LlI 1.' " 11 111,11.'" , ,·,,"v,'rl.,·r lIlI<1 IIli( ' r"l"u("'lll1'"' ('(IId,r,d
The parameter TR can be expected to change slowly due to changes of rotor t(~:nperature, but instantaneously when operating into the field weakening range, when the motor core becomes desaturated. The interaction is described b.Y Fig. 12.20, where it is seen that the flux model ofFig. 12.16 b, controlled by t1w Illc
i:
/11-.
j'( ·(·dl'il,.. li
viI.rinhl<-: : ror /.Iw (·ol)!.ro!.
'l'1)(~
,':iI,illlll.l,,'d ( ·(~ dha('1< :;il';II.t1s C;~iI 01'
270
12. Control of lnduction Motor Drives
12.2 Control of Current-fed lnduction Motor
course only be correct if the parameters of the flux model are in agreement with the parameters of the control plant; this may require on-Une tuning of the model.
271
W
min o' WRe/
1500
" T ' - -1 TR R - 2 C'Ontr'Ol
-l-M'Ot'Or
UsbRel
I I
"~'~Q e~'
I
L'\ UsaReI
"
/ • Flux modeltuned, p' = p
iSb
R'Ol'Or winding sh'Ol1 circuiled, T R = TRO
·1500
o Flux model deluned
W
T 'R • (m'Odel)
min- t
~ ~
e' Jp
o
TRO
WRe/
15'0'0
c:
Q
!I"lrtn~
ai
1
3
" I isa
oI
/
'2
TRO
,c
1-
1
2
(m'OI'Or)
with extemal r'Ol'Or resisl'Or
'2 TRO
·15'00
p'
4
TR
I I
ffi
lR'Ol'Or winding --si '011 circuiled
a
b
Fig. 12.22. Reversing transients of 22 kW wound rotor motor with correctly set and with detuned flux model. (a) Reversing transients with current limit; (b) operating points in TR, T~- plane
The vector diagram in Fig. 12.21 indicates the deviation of the load angle produced by a flux model, that may have been tuned to the motor at nominal temperature and flux leveI, and the actual load angle of the motor = arctanw2TR. The deviation is
o' = arctan w2TR
1'0 c'Onlr'OIIers
FilC( · 12.20. Flux model as an open loop observer for the motor flux,
o
1.1 ... li,·ld orientated currents and the electrical torque
-
.do = o' - o = arctan 1 + W2R2TRT R w2(T
TR)
(12.50)
'
.do > O für T R < T R .do < O für TR > TR
Elevated rotor iSq temperature
(
sd
n)
Desaturation (Field weakening)
"
FIl-( . I~. :lI. VC'No!' clil~J\I" I\ 1I1 01' d nt.lll1c·cI 1111 x '''Cl d..!
li)
caused by overheated motor caused by cold or desaturated motor (field weakening) If the flux model is tuned at rated load of the motor the maximum de viation may cover a range of 0.75TRO < TR < 1. 5TRO with a converter-fed motor where full line voltage starts do not occur. The effect of a detuned flux model is concealed as long as the motor is only partially loaded, because the speed controller would generate the quadrature current reference iSqRef needed for producing the torque, even though the nux may be somewhat increased or reduced. However, when the motor is flllly Jond(~d iLlul <'.;,,/I.,.! ifi damped, there is an actualloss of torque, i.e. the lll (l l.or '" ILIl IIOI. 1)(' flllly load(~d. This is dernollstrated in Fi~. 12.22, where a
22
kW W()llfld 1'101,(11 ' /l1C.l.or wil.h add i l,jo ll/l.l i/l<'l'l.ia
W f U,
cOlIl.roll(·c!
t1lf'(l1lf~h n
12. Control of Induction Motor Drives
272
12.2 Control of Current-fed Induction Motor
speed reversal with tuned and detuned flux model. The tests were performed with two values of rotor resistance and corresponding rotor time constants Tu, [H31, H32). When the flux mo deI is correctly set (1 and 4 in Fig. 12.22 ;~,IJ), full torque is produced, whatever the rotor resistancej with detuned (11Ix model (2 and 3 in Fig. 12.22 a,b), the torque is reduced, resulting in a [ower acceleration. Of course, with an increased rotor resistance the motor opcrates at higher rotor frequency and lower efficiency when producing the saIllC torque, as was shown in Fig. 10.11.
í'sq
iSb
I
ímR/A
ísq/A
RRO
e'jp '
(Temperature)
I I
T RO
ísa
._t.
j'Sd
I
T
I I I
im:~:)::- - t- +i l2..~ -L~
Fl'mRO
i'mR
mM
'I"mR I
-
m
-
- - I
mR)
/a \
aI
5:~
.....
18
\-b
_.
Fig. 12.24. Reversing transients of 22 kW induction motor between field weakening speeds, (a) with constant and (b) with saturation-dependent flux model
MS dRe'
Nonlinear flux model, containing a nonlinear function for saturation
..11','<"1,,, ;,.11<1 a control input for on- line tuning the rotor resistance
~e' úl
'I'hc flux mo dei in Fig. 12.16 b may be further refined by including a 110111
si [.
I mL
Inverse magnetising curve (('I'
1"il',.12.23.
o
I
RR
i"3
n
1/min 2S00
isq/A
I----------------~
i:; 1
273
Speed control/er
Torque control/er m'd
iupar function representing saturation [844) and an additional input for
MS dRe '
a (.( ~ lllpcrature- dependent rotor resistance, as shown in Fig. 12.23 [H32). 'I'lw saturation curve could be identified for a given motor in the factory
prior to commissioning in the field, whereas the slow variations of the I( II.( n' rcsist,ance can be tuned on- line, based on temperature measurements or sOlJle idcntification procedure. An experimental result is seen in Fig. 12.24, wlIcrc t.h(~ sarne motor used before is accelerated at current limit through a wid(~ sp(~(~ d range with a linear flux model according to Fig. 12.16 b and with 1.11<' lIolllinear model of Fig. 12.23j the variation of the magnetising current i,,, 11 i:; also sllowlI ia both cases. Clearly, taking the magnetic nonlinearity illl.o ;Lc'('Ollllt. l'<'HlIltS in a bel;ter trackillg of tlw fillx and again increases th(' lI"lx illllllll lorqu(: at tiw given currenllevel.
e
§
]l §}
8
~
1\ vll,riC'l.y 01' :;r!I<'IIWS has IIc:(:1\ SIlC,J{(::-;(.('c\ for idclIl.il'yiJlg I.hc varyinl', lllo(.or whidl ndl fo I' n(ldiI.iOllH.llllc' ~IS lll'f'lIlC'lIl s hy IIll\,kill ~ II/i(' i.I tt' Ill.n l., li' voll. llf '\ '· <"fi II iI.U", II. (' .r,. IC: ~I .
P:l. f'I' l.lIwlr.('I' '['u, 11i'(ISl. 01'
"r
Field weakening i'mR
Contrai
• :•
Converter and motor
Fig. 12.25. TR- Identification through correlation
An identification method without the need of a voltage measurement [G5, GG) is based on the fact that the output signals iSdRef' iSqRef of the controllcrs iII Ji'if~· 12.25 a.llow decoupled control of the plant in the d-, q_ il)((~0 CJlIly ir I.hc ' ;U ll(\(~ fi lls(~d for the modulation is in agreement with the H.C'I.II :d pCl: lil.iClII I' C1i' I.h,' f1l1x wav('. 1I"II('c' ;1. \ow \cvd dlil.r;\.d.eristic: tc-~st dis-.
12. Control of Induction Motor Drives
274
12.3 Control of Voltage-fed Induction Motor
Lurbance injected into the presumed iSdRef channel must be uncorrelated with the noise in the real isq-channel, expressed for example by the speed error at the input of the speed controller. If there is noticeable correlation it can serve as an indication that the supposed d'-axis is not orthogonal with the actual q-axis and, disregarding other sources of error, the parameter T.k in the flux model needs to be corrected to track the planto The correlation process is manageable when it can be performed in the microprocessor used for controlling the drive. For example, by employing a pseudo-random binary sequence (PRBS) as a test disturbance, the generation of the signal, its cor relation with the speed error and the adjustment of the parameter in the flux model are straight-forward, requiring very little additional computing time. The sign of the cross correlation function points to the direction in which T.k should be adjusted. The sampling time of the PRBS should be chosen sufficiently small to allow the field lag to attenuate the signal, i.e. to avoid an immediate effect on the torque through variation of imR' On the other hand, building up a meaningful correlation function requires sufficient averaging time for suppressing irrelevant noise components. Another RR- adaptation method that is somewhat easier to implement was described in [H31,H32,S42]; it is based On a comparison of volt age signals obtained by measurements and derived from the flux model. Eq. (12.45) may 1>e written in the form (5,5 () t
w hcn ~
= (1 -
(J"
di.mR () R ' )LS ~ = l!s t - s :?os -
(J'
L
S
di.s dt '
(12.51)
RR TR
s 0.200
0.150
0.100
o
s j
5
10
Fig. 12.26. Rw self- tuning employing terminal voltages. (a) Self-tuning scheme j (b) experimental results with 22 kW wound-rotor motor
This scheme is shown in Fig. 12.26 a; to obtain convergence, the signs of WmR and i S q have to be correctly chosen. AIso, there are lower practical limits of speed and load, for which the adaptation yields reliable results. When the rotor current is dose to zero at low load, a correct setting of T.k is not important, provided the parameter is quickly adjusted when load is applied. A test result is seen in Fig. 12.26 b taken on a 22 kW wound rotor motor, with external rotor resistors being switched in and out. The lag of a few seconds would normally be acceptable for correcting a temperature induced variation of the rotor resistance.
c.dt) is the "volt age behind the transient stator impedance". Assuming
/l .,; to i> c kllown, for instance from a temperature senSor in the stator winding, "".! (/ /' S to be relatively unaffected by saturation, ~s (t) may be derived
ri (" II
275
a 111( ~ aSllrement
of terminal volt ages and currentS. When converting this field coordinates defined by the rotor flux mo dei in Fig. 12.20 or nnd for the direct component of the volt age
12.3 Rotor FIux Orientated ControI of Voltage-fed Induction Motor
(·qll itl.ioll lo 1 'L.
'2:1 , wc Co d
.
= (1 -
(J"
dimR . ) L s - - = USd - Rs tSd -
&
(J'
disd Ls - -
. & + WmR (J' L s ts q ,
(12.52) wlaiell reduces in steady state to CSrl
~ 'USd - Rs iSd
+ WmR
(J"
L s iS q
:::::J
O.
(12.53)
tu; I.be angle of transformation p', being derived from an initially detuned JllOdd,
rnay be incorrect this condition is not automatically satisfied but may
1)(' Ils(~d for sdf- tuning the model according to
,, /",
(I(UII;Rj{u) _ CS d II - fi,. ("
, ''lf l
(12.Gt1)
The field orientated control scheme in Sect. 12.2 was introduced with the sim plifying assumption of current sources in the stator circuits; this was achieved by fast cOntrol loops for the stator currents, making use of MOSFET- or IGBT- converters with ample ceiling voltage and short access time due to high switching frequency. Of course, these conditions cannot be met at maximum speed, where rectangular volt age waveforms are desired for fully utilising the converter; hence the concept of impressing currents becomes impractical in the field weakening speed range. Another area of application, where the approach with current sources and, in steady state condition, nearly sinu soidal stator currents has to be abandoned is that of pulse-width modulated voltage-soUl'ce converters for larger drives employing thyristors and GTO thyristors. Tlt e fiwitching frequency of such converters is usually below 1 kU'l" a s colllJlar< ~d t.o 10 kHz for IGBT- or 20 kH'l, for MOSFET- converters. A s ;t. C ()Jl :; ('(l'l('ll C( ~, t1w ltanllOlli('s of t.lw :;t.nt.oj' nllT('llts may bc af apprccia 1.1(, 111 111',11 \ 1.11.1" tl.lld 1.1)(' lI'ifillllll't.iotl 01' II<':t.t'I.v "ill :: I. IIIII.i lllCOll :;" (·. \lrl'( ~ llt ('.o!ll.rol 1"'("1111" :: qll' ·l l l.i"" II.loI', ',
276
12. Control of Induction Motor Drives
12.3 Control of Voltage-fed Induction Motor
As a result, the stator voItage equation (10.50) now has to be taken into account as well; by inserting the magnetising current vector, Eq. (12.22), representing the rotor flux, Eq. (12.45) results. Converting the volt age vector to field coordinates as in Eq. (12.26), Y,s (t)
e
-j
(2
=
USd
+ J uS q , •
277
Us3
(12.55)
and splitting in real and imaginary components yields (JTs
di + iSd
disd
=
USd Rs -
(J T S
. dis q di + ZSq
=
uS q ( ) ' Rs 1 - (J Ts WmR ZmR -
(1 -
dimR
(J) Ts
~
.
+ (JTSWmR ZSq
,
. (J Ts WmR ZSd .
(12.56) (
... ...
12.57)
e sd
,, ------f-- ,,
Rs
These equations describe the interactions between the field-orientated sta tor volt ages and currents; together with Eqs . (12.27, 12.33, 12.34) and the equations for the mechanical part of the plant (10.52, 10.53) they form the mathematical model of the voltage-fed induction motor in field coordinates. This model is contained iu graphical form in the upper part of Fig. 12.27. The additional right-hand si de terms in Eqs. (12.56, 12.57) constitute undesirable coupling terms which could be compensated by suitable feed-forward signals aL tI.e output side of the current controllers [G3, G4]. The principie of field orientated control, applied to this scheme, is shown iII I.h(! lower part of Fig. 12.27. With the assumption that the control dynamics 01' I.II
+j
USq)Ref
ej
(2'
;
(12.58)
I' I: : ; 1I ~aill the angle of the fundamental flux wave as determined by a flux iI,('([llisitiOll scheme, for instance the flux model shown in Fig. 12.16 b. The Jllaill dilrcrence between the control schemes in Fig. 12.14 and Fig. 12.27 is l.h;tI, I.he current control is now performed in field coordinates, where the con I.roll(!rs are processing DC signals in steady state, thus avoiding the problem or pltase shift of AC current loops. AIso, the digital part of the control may 1I0W illclude the pulse-width-modulator and can be extended down to thc I~(!ll(!ration of the switching signals for the converter. Th(! rCllIaillder of the control system is similar to the one shown in Fig. I :~. 1,1. TI.e ~tructure in Fig. 12.27 is somewhat redundant since the task of is d - 1Llld isC/-mutrol could also be assumed by the flux and torque controlknl, (lrllil.i.illg th(! CIIIT('1I1; COl1t,fo\l('rs. Field weakening is adricvctl with a fllllCt.ioll r:llIlt'J'ill.or
I" ,
I\ljll lI lI .I'
()t l/atl hl .. .
Flux control/er
Contrai
US1RfJI
,
US2R(Jf
,E
U 3Re'
s
. ~ :"1.:.:.:.:.::: ~.
L-J-~·F::::·~
~L-:'-ó4-. (J)Rel :-------.: €Re/
usqRef :.:...•.."..:_-.:
Phase spllt
Coordinate transformo
Current control/ers
Speed control/er
Angle conlrol/er
Fig. 12.27. Block diagram of induction motor control in field coordinates, assuming volt age source converter (- - - optional)
A 22 kW induction motor fed by a voltage source transistor converter with vectorial PWM and controlled by a signal processor programmed in C language has been designed and tested in the laboratorYj the somewhat modified control structure is seen in Fig. 12.28 [K52]. Field-weakening is now performed by a voltage limiter, which corresponds to the solution shown in Fig. 7.13 for a DC motor. The converter was a 30 kVA experimental unit with 540 V DC link voltage and a clock frequency of 1.37 kHzj the sampling frequency for the inner loops is twice the clock frequency, Fig. 11.8j an integer fraction of this frequency was used for the outer control loops. A reversing transient of the speed controlled drive is depicted in Fig. 12.29, with the inertia of the drive increased by a DC dynamometerj it confirms the expectations by showing nearly perfect decoupling of the d - q- axeSj I.\tis is manifested by the constant magnetisiug current and the rapid rise 1.0 lll axilfllll/1 value of the quadrature current representing torque. While the s tator Cllrrcllts cxhihit large transicnt tenns, the magnetising current remains pracl.ically :; illll~;()idali thiti is dll(~ t.o til<' Iag TR, acting as a low pass filter. 'J'!H! J'i ~ (' I.illll' "r /.III' I,o('qlll! of a hulIl. '2 1/1.8 COI'fl'spOllds to the limits imposed I,.\' I.h, · (',,111.1'.. 1 d'y II fi lII i (' H 01' /.IIC' )'WM ('OIlV I 'I'I," r; I.hi s rnpid I'ns poll:w is HIIP('
12.4 Induction Motor with a Current Source Converter
12 . Control or Induction Motor Drives
278
S/a/or coordina/es
Field coordina/es Flux
Vol/age limi/er
PWM
e jp'
mM R., úl
I ImM
'",R"
lusaR,,11
I
I~
t.I,,~ "'"'1"'111.
0.6
0.7
, 0.6
: 0.7
, 0.5
, 0 .9
, 1.0
, 1.1
,I/S
1.2
..
~,_ 0.8 0.9 1.0
I , 0.4
, 0.8
I
,...
,tis
I
1.1
1.2
ImR
0.1
0.2
0.3
o C/\J0.vC\ P\1\\A74
-,
iI.lld volt:q'.'· v,·,·t."!'!:
, 0.3
100t A
u
0.4
0.5
0.6
0.7
0.8
0.9
1.0
ti s 1.2
1.1
Isa
70\ A f\
f\ ){\,
I
f\
Í \ {\\I ( ' \I \ I -"/\'1
n
yA
f\ 1"\ A \/~}\r\t
'\f
\./ f\ q
ti
s
3~p Af\ f\ f\ f\1JLf\ fV) '30VV
rior to that of a high-performance De drive with a 6-pulse line-commutated converter at 50 Hz. The pertinent loci of the stator- and the magnetising Clll'rent vectors i.s (t) , i.mR(t) are seen in Fig. 12.30. Again, the magnetis iIlg cmrent vector reverses on a near perfect circular path, while the stator ('11 rrcllt vector is temporarily boosted to produ ce the desired torque. 'r1t(~ ulock diagram shown in Fig. 12.27 is simplified in so far as the internal !'c-('dh;lCk signals seem to be derived directly from the motor, while in reality I.Iwy are gcnerated by the flux model computed in parallel to the motor, I"i)',. 12.20. This observer-like structure is shown in more detail in Fig. 12.31. 11,'(((',(' OIW should distinguish between the real but not accessible variables, ('or illstallce p, and the model-based approximate signals such as p'. Since the coordillat.e transformations at the input of the converter and in the flux model ilJ'(~ l)(~rformed by the sarne angle p', the actual stator frequency Wl is in exact ll!-,:n~e1l1eIlt with the frequency WmR generated in the flux mode!. ln steady ~; I.a(.c~ it corresponds (assuming a two pole machine) to the extrapolated no loatl speed of the motor; if the measured speed signal W is correct, then the csl.jllli\.ted rotor frequency is also in agreement with the actual rotor frequency, C,)~~ = W2 . Of course, should the model parameter TR be detuned, the flux could (kvial,c~ from the desired value; at reduced flux, the rotor frequency would be 1.00 high alld vice versa, but the torque would still be correct, unless tlH~ CIII,(,C~J)(.S an ~ liJllitcd , ot.herwise the speed controller (assuming it to be frce of dri!'t.) illil.ia(.c ~ :j lhe Ilece::;sary changc~s. lu t.hc~ l"wl~r [larl o[ Fig. 12.31 lhe Ollc.lilll ~ Uwt.ulIillg scheme is HPI~ll t,hn.t. W lu< d" ::l'I'i[,,',] iIi ro'il';. 12.2(;; ah;o ali all';llI'it.lllll iH illdi(';,i.< ~ d for c·slilllat.il\/-'; I.ho IlI'l.w( ·(·u
, __ ., 0.1 0.2
~
sq
t ".
0.0
Fig. 12.28. Microcomputer control or induction motor wilh a voltage source PWM-converter
I)
O~
:1 , , , , , , , , , , , ,
.~
.'1 111'.1(' ,/ .
, 0.2
tA
i",
-jP'
l
úl
I
e·
100
O
u VQ
t
, 0.1
,600
Converter
m Speed con/rol/er
Jmin'1
60~
279
iII
Ht.c'ady Ht.Il.t.c;
II: :
ImRa (V)JV\ fi f\1Jlf\ f\ f\JJLf\ tis WIJ2J VrJJ V O.4 ~.6 VoV VoWíJ V}9V IJ1YJ Vlj,1V V112
Fig. 12.29. Reversing transient or 22 kW induction motor drive with voltage source converter and signal processar control at no load iSb
imRb
Current Iimit
100
t isq
/
100
75 imRa
50 25
j
O 30 -100
a
b
.I 35
imR
C
Fig. 12.30. Stator current and magnetising current vectors of PWM-converter drive during the speed rever sal shown in Fig. 12.29 (a), (b) Currents in stator coordinates, (c) in field coordinates
shown in Fig. 12.32 the needed signals are available in the control computer, where they are generated with the sarne transformation angle p'. With the help of Eqs. (10.72,10.76) this results in ,
7r
, ( 1 - (J) W2 T R ( T )2
tau (I(! -I- :-) = cot (-
1 + (J
W2
n
(12.59)
1'1'01" UI(' ... ·:ltdt.i/ll·; C[lIaclrat.i/: cCfllat.ioll for (w'J Tu)' ltml the known va.hlP HII.,U", !' (·lI t, i ll uLl./· (lr t.lw 11ICldd t"l.Il1.l1l1'l.(·r '/';( .. .,,,1<1 hc~ dC 'rivc:c1 ,
(Ir w~\
280
12.4 Induction Motor with a Current Source Converter
12. ControI of Induction Motor Drives
u' - R i' SqRef S Sq I
i'Sq
... ('!sRef-RS1S)e-
~/-
Sd
u
Induction motor
/ /
Control
Coordinate fransformation
L
U~dRe/,------- -,
UsaRel '----'
-
~
U s3Ref
Phase sp /il
w
p'
iSd
iSd
12.4 Field Orientated Control of Induction Motor with a Current Source Converter
= is(t) ej((t) = V3iD(t) ej((t) ,
(12.60)
i'Sq
iSb
w
RR
e· jp •
RRO
Flux model
is.
1'4l t,
0
/ Ref - RS
is e-j
Fig. 12.32. Computing the steady state phase angle cp' from measured voltages and currents in field coordinates
is(t)
MOil$Urements
1... .
L
The current source thyristor converter described in Sect. 11.3 is of interest for many AC drive problems because of the simplicity of the power circuito As shown in Fig. 11.14 the control functions are naturally performed in po lar coordinates, where the magnitude of the stator current vector, being in noncommutation state directly proportional to the link current, is controlled through the line side converter, while the angle ((t) of the stator current vector,
Us2 Ref
0
.
P'
~
usRs!
O~
\'
j
281
Measured speed
mM
1.1.;1. ' -~
f('f
mR
imRO ) ('f'mRO
)
r.=~_---,I (Seturelion)
~
- Tuning of flux model
'f'mR
len ('II' H 12)
i',q~ ,i'5d ~ guadr. Equ.)-----+ (W 2 TR)' usdRef ~ ==== U'sqRel
Estimation of phase angle
."i,.t.
12.31. Signa l How dingram of inc\uctioI1 motor drive wit.1l ,;;t!.lIr;d,ioll depelll.lcllt fll.l x JIIoucl ;wd on,.linp. tnnillg 01'
Ru
is determined by the switching state of the machine side converter; in view of the link reactor and the reactances of the motor, is(t) and ((t) are both continuous functions. Outside the commutation intervals ((t) is identical with the switching state ç(t) of the machine side converter, where ç(t) advances with increments of 60 0 in forward or reverse direction. Control of ç(t) mod 27r could be accomplished by switching a hardware or software ring counter with six states, To this basic control structure which is motivated by the power converter circuit, the principIe of field orientation can also be appliedj one possible solution is drawn in Fig. 12.33. ln the upper right hand comer it contains the modeI of the induction motor in field coordinates assuming impressed stator currents. The direct and quadrature current components iSd, iS q are obtained from is and the load angle 6 = ( - e by a polar/rectangular conversion (P/R); the inverse operation (R/P) is seen at the output of the flux and torque controllers. The transformation into stator coordinates is achieved by simply addin g again to the load angle reference 6Ref the angle e mod 27r of the ftnx wave, a~ obtained by sorne mcthod of flux acquisition. The output sigllaIs 01' t.Il1~ d riv(~ cOllt,rol an~ j;h!: ref(·!'(·u\'.c valllcs for the magnitude and ;I.llf~\('
282
12.4 Induction Motor with a Current Source Converter
12. ControI of Induction Motor Drives
Induction motor
283
of the motor reactances. Clearly the current source converter is not best suited for continuous position control, where extended operation at very low speed or even standstill with controlled torque may be required. The situation is different on discontinuously operated drives with position control such as needed for boring machines or screw-down drives on reversing rolling mills; in these applications the positioning drive is made inoperative and the brakes are set, once the desired angular position has been reached.
I
Converter with current contrai
Microcomput~ Converter
8086 iS Re' ~..J -3 io Re'
imR
mM
T
I I I
DIA
and motor
is
'Re' + P ~ ç Re'
imR
I
mMRe'
mM
Fig. 12.33. Simplified controI scheme of induction motor sllpplied by current source converter roAs!
iII Fig. 12.33 in simplified formo TD is the filter time constant of the DC link circuit, UD the effective supply volt age of the machine side converter which acts as a disturbance to the current controlloop; UD depends on speed, flux alld stator current of the motor. A problem arises in the second control channel where a finely discretised rt'f(~rence angle (Rei is to be tracked by a feedback signal having only six discrete angular states. An approximate solution is again to employ pulse widLlt modulation, i.e. switching the ring counter back and forth between I.wo adjacent states [G5, G 15, W9]; this may be achieved with the help of an :uq:{lc controlloop. For this purpose the state ç of the ring counter which is rcadily available in digital form can be used as feedback signal. The current augle ( follows the lead from the ring counter with a commutation lag that is rdated to the leakage reactance of the motor; the time required for commu tation depends on the link current because at reduced current it takes longe r (.0 recharge the capacitors; this is qualitatively modelled in Fig. 12.33 by a current dependent gain factor. The reversible switching of the ring counter t,akes place only at low speed, it ceases as the stator frequency rises and the ilugle controller reverts to a unidirectional progression. The pulse-width modulation achieved by the angle controlloop has the nJdcJ benefit of reducing the effect of torque pulsations at low speed which are ca\ls(~d hy the interactions of the stepped stator ampcreturns--wave anel Llt(~ eOll(.iullomdy Illoviug flux wave. However, as was meutioncd Ldore, tlLP 1'1'( ' <111<'II<:Y (Ir (.\I<~ plllsl~- widLh Illndlllat.ioll is 11111<'11 lowI:r t.!t'UI iII til(: ca:'H~ of :L volt.;q·,(· ::"111"'" '·"/lVI'rl.l'I', wlwl'(' I.ll<: (,Ollllllll(.;t.!,i,," ii: plact.il'.lllly ia.!I'p I'lldl'lI(.
ro
DIA
I Fig. 12.34. Microcomputer controI of induction motor with a current source converter
The speed control scheme of an experimental 22 kW drive, employing a commercial current source converter and an 8086 single board computer is seen in Fig. 12.34 [G5, G6]. The main portion of the block diagram is similar to the one shown in Fig. 12.14; differences are evident in the output section, where an analogue reference for the link current and a pulse sequence for controlling the ring counter are now generated; another addition is the feed-forward signal to the angle control loop to reduce its velocity error. It is noted that the converter could be controlled at the lower leveI by a hardware sequencer which maintains minimum separation of the firing pulses to assure complete commutation; these protective functions are not shown in the diagram. At an advanced stage of development they can also be realised by s()ft.wa.r(~ or special integrated hardware. /\. J'('VI'I'SiIlH (,rallsient. of t,!te 22 kW drive a.t half ratcd speed is recorded in Fi,,;, I/..:S r, wltil'il 1I.1~lI.ill sltows !.II(! !'.nlld d(~(,()llplillf ~ of (,lw d - q-a)(:s adli(~vl~d
284
12, Control of Induction Motor Drives
00/
12.4 Induction Motor with a Current Source Converter
400 min- 1
750 min- 1
-75~
~"""'.'.
i
285
~
~~
•
-400 mín- 1
Sq
010'
,
annn. rri r(ii
rt ,nl II'
iSa
n n~!~ b1l.'AWA\Y.\WtPmv m"
1I1Ittft""\lnn U1J rOTu fi n
ft I IilldfUJ
n~nnftft.ftnft'r O OH
I•
1"\
I
I
I
iSa
I.IIINI
•
t1 h A,nllnA,nll ~ II
,
..
imRa
~JJJAMMA
II
l\AAMMAAAAAAMMMAMAAAAMAAAA .. t
1 s -----+l
.~~ M AAMA~~JMMM~~~~i~~~M~~JA~A~j"
AAAAAA /\ I~
1s
mRa
i
1\ A/\,~,~AP,A
lo
~I
ç
4AAAAA A ~ r\~~~~~t~ •t Fig. 12.36. Slow speed reversal of current source converter drive
"'ig, 1 2,35. Reversing transient at half nominal speed ",. a, :2:2 kW current source converter drive
"y li('ld oricntated controI. The rise time of the torque is about 10 ms, the 11I;1./',lIil,lldc of the flux is maintained nearly constant. As seen in the lower i,1':t('(',~ till' lllagnetising current is practically sinusoidal while the stator cur r('ld,~ have a highly distorted waveform. The motor was again connected to a I)C dyll;uilorneter at no load. Tlw lH!haviour of the drive at low speed is demonstrated by the transient íll I"il-(, 12,36 where a slow speed ramp causes the pulse-width modulation i,o I>ccoJlle effective around zero speed. This is clearly seen from the trace of 1.11(: swi(.ching state ç(t) of the ring counter; as soon as the speed leaves the Ilarrow band around standstill, the modulation ceases and the angle controller colltilllWS with a unidirectional switching sequence. (;lIar;Lct.eristic loci of the current vectors are seen in Fig. 12,37 during a :; p\'cd rcversal at very low speed, Initially the stator current v~ctor is stepping :1I'OIlII
The locus of the magnetising current vector, Fig. 12.37 b, is again hardly affected by the switching operation, even though the radius of the lo cus is not quite as perfectly constant as in the case of the volt age source converter drive, Fig, 12.30 b, indicating a somewhat reduced dynamic performance; this is of course due to the lower switching frequency of the current source converter. ln general it can be stated that both types of thyristor-fed induction motor drives show similar behaviour in steady state operation and for large ampli tude transients. Differences do exist, however, at very low speed, where the current source converter causes a slightly cogging motion while the volt age source PWM converter drive operates smoothly even at standstill. A note worthy feature of the current source converter is that it performs with very little audible noise which is in contrast to some voltage source PWM convert ers unless a higher switching frequency beyond 16 kHz or special modulation techniques are employed, Further progress is achieved with the circuit in Fig. 11.16, where a current source PWM-converter with GTO-thyristors is supplying an AC motor hav iug filtcr caIH\.('il,ors across its terminais [A18,A19,N17,N18,N19,N21). With a ~;lIii,al)le <"Iloi('(' 01' 1.1\<' capacil:ors al1d lhe switching frequency this results iII 1',II';d,I,I' illl!ll'(lv( ,d wi\.V(:fol'lm; 01' 1.11(' 1l11li,OJ' volt.aj';('s aucl currents than is
286
12. Control of Induction Motor Drives
287
12.4 Induction Motor with a Current Source Converter
~~/'-------- Currentlimit
es () t = (1 -
-
(J
dimR )Ls - - = (1 -
dt
(J
mR ) L s [di --
dt
. ] + J. WmR ZmR
e j{! .
(12.62) This is now supplemented in Fig. 11.16 by the stator current equation
~---~
~
.
1~'
\;/-\
d'gs . . Zs, C -dt- = -zp - -
+~R(t;
-
(12.63)
where i p is the current vector supplied by the GTO-converter,
ip = V3iD(t) ejW ) ,
No load current
a
b
Fig. 12.37. Stator and magnetising current vectors of current source converter drive during a speed reversal at 200 1/min
possible with either volt age source or current source converters discussed so far. As a consequence, the additional copper- and iron- losses in the motor are reducedj there are also lower volt age stresses on the insulation of the willdings. Sillce the converter is now feeding a low impedance load, the commutation al',;~iu l>(~comes independent of the motor, as with a voltage source converter, i tlld ('ali operate at a comparable switching frequency, producing a smooth 1II"l.or Lo[que. Instead of zero volt age intervals there is now the possibility of "lllpl"yillg zero current intervals where the link current bypasses the motor, I.J 111:-: (' n ~ al.i llg additional degrees of freedom for designing PWM strategies. At 1,11<' ,::ulle time, the advantages of regeneration by inverting the link voltage i I d ,( '
(12.64)
and is = is(t) ej((t) is the vector of the stator currents. As with a volt age source converter, Fig. 11. 7, the angle ~ (t) of the converter current can assume six discrete equidistant valuesj in addition there are zero vectors i p = O caused by the short circuits of the DC-link, when the link current bypasses the load. The block diagram in Fig. 12.38 describes the extended dynamic system, using a hybrid representation of the motor, where the stator side quantities are left in stator coordinates, while the remaining motor dynamics are in field coordinates. The block diagram of the GTO-converter, in accordance with its mode of operation, could also be drawn in polar form o
§s
(I)
I eip
esb
, IpJ
esq
(i)mR
iS J
-:S~
"'ig. 12 .38. Block diagram of current source PWM converter with GTO-thyristors
(7
'I' . ,. _d'i s + '/., . ='lli.s - - (1 . dI. ~,..,
ns
-
) T ,, di mH (J - =lis- -~8 -
.
dt
Rs
'
Rs '
(12.(j L)
UI(' vII]I,IIW ' t':: i:,; illi.<'rpr<'l.t'd a!-i 1.111' illlllll'('c1 vIIll.a l(n "hdlillcl ' I.lll~ trnll hk llL llllJ)C'llll.lI( ·('" lIf UII' IlloL"r,
",,,I
(,;l. p;lC : i~i ve
f'ilter
1\:; 111(:lIl.iol\e l! ), (' i'ore, t.lle (:olltrol of I.hi H I.ype (lf drive I t'lIl1irc 'lI
It
l d l ~ h 1"'1'1'01'11111,11('(' Illft ll a l 1I1"""('flHo r L.,
,,('hit'vP
i~ quit.e ;1 f;l,ahlf'
complcx, alld w"l1
12. Contrai of Induction Motor Drives
288
12.5 Control of an Induction Motor Without a Mechanical Sensor
"~~~, i S8 , b lA
20
oV/\\ I/ I~\ /\
11\ J
'
Fig. 12.39. Waveforms of terminal volt ages and currents of a 55 kW induction lIIotor with current source GTO converter operating at a stator frequency of 50 Hz. It.cactive power of the capacitive filter is 24 kVA, switching frequency of converter is SOO Hz
F ig . 12.40. Step response of torque contrai loop with the 55 kW current source (:'1'0 converter drive used for Fig. 12.38
dalllJ)( :d response. Some steady state and dynamic results, that have been nlllairred with a 55 lcW motor using a model based predictive control strat "['S Ii\ 18, A19], arc seen in Figs. 12.39 and 12.40. Clearly, this could be a very "roll)i~;iIl1!, approadJ to future AC motor drives, eombiniug smooth wavefonns 01' voILaf!;<'s alld CUlTf:nts with cxcellent dynamic pp.rforrnallc
ah;() \)clldicial for lhe d(~ sirr,1I of t-hc l11oLor, IH:CillIS<~ (,hc c\
289
ages caused by the "hard-switching" converter are avoided and the motor is relieved from the task of serving as a filter choke. At higher power ratings sueh as for low speed piston eompressor-, mine hoist- or rolling mill drives, cycloeonverters may also offer a solution as long as their frequeney limit is not exceeded. The control is similar to the one described in Sect. 12.2 where control loops for the stator eurrents provide nearly impressed and close to sinusoidal currents [H28]. If the range of stator frequency rules out the use of a cycloconverter, a synchronous motor with a load commutated current source converter could be used; this is diseussed in Seet. 14.3.
12.5 Control of an Induction Motor Without a Mechanical Sensor 12.5.1 Machine Model in Stator Flux Coordinates After the basic problems of controlling induction motors with high dynamic performance had been solved by applying the principIe of field orientation, where information on the magnitude and orientation of the flux wave is de rived from a "flux model", i.e. a mathematical model computed on a mi crocomputer in real time, the attention of researchers has turned towards simplification as well as refinement of these quite sophisticated control meth ods. One point where further progress was felt to be desirable is the omission of the mechanical speedjposition sensor needed with many of these control schemes; for instance, the rotor-flux model introduced in Fig. 12.16 b calls for a measurement of speed, even in cases where the specification of only a moderate accuracy would not make a speed sensor mandatory. While elec trical measurements are usually acceptable since the sensors can be placed anywhere, preferably inside the converter cabinet, a mechanical sensor is of ten considered a nuisance because of space restrictions or the added cost and complexity, particularly with smaller motors. Of course, a certain loss of ac curacy and dynamic response must be accepted when an estimate of speed serves as a feedback signal, hence such schemes may not meet high perfor mance demands, for instance as required by machine tool feed drives, where scnsors will be inevitable. One reason, why the omission of a mechanical sensor has become such a prominent topic, is surely the fact that the powerful microcomputers now used for the control are applicable to the estimation of not measurable internal ql1antities; hence it makes sense to employ their potential also for substituting L1u: mechanical scnsor. Had microcomputers been available for controlling DC 1I1;u:lrill<:S, t.Jw sallW demands would have emerged there. '1'1)(' Vil,ri()ll:; pl'O[>()sa]s for thr dCl:iif':n of eontrolled AC drives without a IIIC'c1 11 l.l,iC'n.l HC' II !'. c>r hav,' iII COllrlllOIl Um!. ol1ly l.crlllillal <[uantities, i.e. statOr 1'0)1 1.11./',, '11 1I.1le1 "111'1"'111./: I I,/'i ° 1IIC'II,;·;ur('eI !"r' ()11I ""llidl UI<' illforrllaLioll 011 fi,,:, aJl
12. Control of lnduction Motor Drives
290
12.5 Control of an lnduction Motor Without a Mechanical Sensor
speed of the motor must be derived, based on a nominal knowledge of the important motor parameters. An ideal estimation algorithm should produce the necessary signals on the flux wave as we11 as mechanical speed and, in addition, provide information on changing motor parameters, so that the flux- and speed- mo deIs remain tuned to the motor. Of course, it is unlikely that a11 these expectations can be fulfi11ed at the sarne time, still it is a goal worth pursuing and, when applying powerful numerical procedures, there may indeed be effective ways for solving some of these problems, e.g. [78b, H46, H69, H72, H73, 05, 010, T4, 62]. The three tasks mentioned may be summarised under the headings "Ob servation" and "Identification" • Acquisition of flux wave with regard to magnitude and angular position, • Reconstruction of speed, • Estimation of machine and load parameters; they are posing different demands with regard to response time as we11 as accuracy. The estimation can only be based on the available mathematical mo deI of the motor, a set of six nonlinear real- valued differential equations with uncertain parameters, Eqs.((10.50-10.53). Since flux acquisition is the most time- criticai, being a basis for the torque control, it is reasonable to select a coordinate frame of reference that can be directly employed for the subsequent control; the choice is either a stator tlux- or a rotor flux- based coordinate system. Both have their advantages iLlld disadvantages as discussed, for example in Sect. 12.2 and 13.1. Since roLor currents cannot be measured with cage motors and a11 measurements have to be stator based, a stator flux coordinate system related to the equiv al('lIt circuit in Fig. 10.7 c appears interesting [D26,K74,S63]. However, the '1lwstion of choosing the coordinate system is some times overrated because I.ll( ~ difference between stator- and rotor-flux is leakage flux that cannot be acclIl'atdy modelled anyway, Eqs. (10.25, 10.32). Still, it seems interesting in ví(~w of the stator based measurements to try also a stator-flux orientated ;qlproach.
'1/) (t)
= Lo lmS,
(12.65)
where ~.m8 = ims( t) ej l'I ~ [> r ese llts
p.(t)
= (1
+ (J's) 15 + IR e j <
(12.66)
a stator based magnetising current, the unaccessible rotor curren[.
IlIay IH' dimin at.crl frolO Eqs.(10.50, 10.51), resulting in two equatioIls for ~"'s
ali" i:;,
ti"/'.' ti, .,'
,li", ....
Lo.lI.
'II" ,
.'
U.-; is
and (J'
Ls d!:
+ Ls (ljTR -
jw(J') is
= Lo d~~s + Lo (ljTR -
jw) i ms . (12.68)
The "induced voltage behind the stator resistance" is integrated in Eq.(12.67) resulting in ims and /.L. This is the moving frame of reference defining "stator flux coordinates"; the stator current vector in field coordinates is then .
'!!.s e
-j p . '
= tSd
.. + JZSq
(12.69)
.
When separating from Eqs. (12.67, 12.68) an inner voltage vector which is defined by measurable quantities,
Y..s =:!!s - (Rs
1 + (J's. + 1 + (J'R RR) 15 -
(J'
dis LS dt
'
(12.70)
the motor speed may be estimated from
. i ms . ) JwLs ( - - - - (J'1s = Y..s 1 + (J'S
RR . + -1-+ (J'R 1mS .
(12.71)
Substituting the rotor current in Eq.(1O.52) by i ms , yields the motor torque mM (t)
= ~ Lolm[is 1:"51 = ~ Lo ims iS q .
(12.72)
These equations still correspond to the original model of the induction motor, Eqs. (10.50-10.53), only now written in terms of stator- flux coordinates; they are graphica11y represented at the right hand side of Fig. 12.41 with the measurable terminal quantities:!!s and is in stator- and the motor dynamics in stator flux- coordinates. 12.5.2 Example of an "Encoderless Control"
With the definition of a stator flux vector - 8
291
(1 2.(;7)
Proceeding in Fig. 12.41 from the motor terminaIs to the left, a voltage source converter is assumed, feeding the motor with two orthogonal AC source volt ages USa and USb, which in tum are controlled by command signals USaRe! and USbRe!; the converter is represented by its effective delay. If the reference voltages, available as digital signals in the control computer, can be substi tuted for the terminal voltages, a difficult measuring problem is avoided, because the voltages produced by a voltage source converter have a wide bandwidt:h, calling for costly AjD converters; another option worth trying is the I'I~COllstrllcl.ioll of the terminal voltag('s from the DC link voltage aneI swit:chill/~ si/';II;dH, t,nkiug Uw protccl.iVl' illt.(~rva.l H iuto account. TIl(' 1';\,It 1'111'1.111'(' to til(' I"fl. 01' I.h( ~ d illg r n lll r ('I)J' ('S (:llls itU cxalllple 01' iUl " ('IICt"I"I'I" HM
" 11 11\.1,.. 1" ;
í l.n illpllt.li
111'1'
UI<'
11 11'/1/1 111".\
,'I( ·(·t dn d Iii ".IHtlt.: iJldi"Il,l,,'d
12, Contrai of Induction Motor Drives
292
12.5 Contrai of an Induction Motor Without a Mechanical Sensor
2o
::>
~I
'--
by the orthogonal set USa, USb, is a, iSb of AC quantities. Integration of the "voltages behind the stator resistances", USa - RsiSa and USb - Rsisb pro duces the stator fiux; transformation of Eq. (12.67) into field coordinates results in
JI~~
>< :::,C ::::: t b
~c: C(~ t~-, -y- ~' o 0
---
0 QJ
dims
Lo ( ----;]i""
~
~
{l
dJi dt = WmS
o E o'-
C(~I ,
.S ü .c c:
~
:
-o
õ(ij
1~1ii
E
,
___ -
.,.,
C(
'-,
o~
I
(3'"
n
Vl
.
o~
=>
, '"
1 __ I'" + : +
:~
-~:, ~~
,
(1--o--.t. 1 ~
I
~
I ~ I L r - - __I.J
,, 'ti ,, o'" ,
t
" "
o~
Vl
,,,
r:r.. , o'll ::>
-~I II
,
,
::>r ________
C:r:..
-~ -~~:,
o ~
I)
11 .
dt
..s 'i-: '",
o~
:>
:t)~I~
~
-
t::
~
-a :::.
~
dJi-
Lo -
~-~
Ui : Cc QJ , Cc
(13'-
!f~
(12.73)
.
(12.75)
o'"
--.I b
~~
c:
dims dt
~I~
;J
QJ<1l -Cl .~
CJ)CJ)
o
Rsisq) ,
(12.74)
Lo - - = USd - RStSd ,
I~~
"O CJ) QJ
=>
~
+ j(usq -
is the instantaneous angular velocity of the stator flux vector. Hence, the equations of the stator flux model are
Õ
~
\I)
( . ) _j = 'Y!.s - Rs!s e I-'
where
<1l -Cl
(\)
.. dJi-) + JZmS di
= (USd - Rsisd)
CJ)
'::J Ir)
293
o~ ,
;-~
___ __ ___ t j
- -.:
UO!jt!W!IS8
:
---4 ----,..----1 p6SdS I VI
_E
..
- -g: _ 1:l :
~
~b
,--E
lo'"
"'--.I
II
I
<1l QJ , E QJ ,
~Q:
~
0
8
.,... '-'
~1:J
~:rll
-----'=---- - -
=
ZmS
(12.76)
Deriving the stator flux from terminal voltages and currents is difficult be cause the open integrations at low stator frequency are subject to sensing errors and integrator drift; on the other hand, if the operation is performed inside a feedback loop, as is the case when being carried out in field coordi uates, the effects of integrator drift resemble normal offsets, common in ana togue signal processing. The result of the flux modelling are estimates of the magnetising current vector expressed by i~s, Ji-', and w~s' Clearly when mea snring stator voltages and currents, the flux model does not require a speed signal; this makes the scheme fit the needs of "encoderless" controi. The flux lllodel shown in the upper left hand part of Fig. 12.41 may be interpreted as lhe inverse of the corresponding part of the motor dynamics, creating a copy Dr the stator flux; its function resembles that of a phase locked loop (PLL). 'l'l w signal processing may be done with an analoguejdigital sensing circuit, whcre the trigonometric functions are represented digitally in a ROM, to be 1I11llti plied in D j A converters with analogue volt age signals [JI5, JI6]. Once estimates of the magnitude and orientation of stator flux are estab lislted, it is easy to compute with a microcomputer the remaining quantities Iwcd ed for a two- channel field orientated control structure. This is seen in the lowI~r left ha nd part of Fig. 12.41, where a voltage limiter generates the flux 1('''I ' I'(~JI('.I ~i :,, 8 J1cJ as a command value for a flux controller producing i~dRef ; :.1111 ihrl ,Y, I . ll( ~ (lJI lpllt of Lhe speed cont.rolkr is the reference for the quadrature ' · IIII ""ld. , ii.. I ' J" ' I ' h( ~ stator cml'('Jlt ('.()lIl.rol iII (idd coordin ates corresponds Lo 1.11 11.1. :-l huwlI iII 1"IJ':. 12. :2'7. Ckarl,V, ir 1.11\.' voll.iI.,',":' '/I~, : (i.) r.outaininfJ; rd pvallt, 111 11'('1'i11 1J !.iI ", 0111.111' 101.1" 1'11.11 ))(' III'IIVI'" 1'1'11111 1.111' '.I ' I' l1l i llil.l qll ll.lIl.it.íl'H, I.lIn(' ,'11
"'l1{ . I ~ .ill , 1').I' rtlIUJ lt- 1I'l odd .. r /lll AC IIl OLIl r jfl III.!Ü'lr IIII X ' :'I'II'diflnt l'lI ll lld (' J(/H I ~ I! I(' III" JI t"tln h ',,1 t!dU'llu ' Wit.lHHll. J\ uW l' hunl",d iI(~ III U II'
uSq - Rsisq
--'--.--~
II'
294
12. Control of Induction Motor Drives
is a good chance of obtaining estimates of the motor speed w' and the rotor resistance Rk. Reconstruction of 1Ls with the help of Eq. (12.71) involves a differentiation of the stator currents which can again be implemented in analogue formo When transforming Eq. (12.70) into field coordinates
1Ls
e
- j li-
= VSd + )VSq = -RR OmS + wuLsis, + jwLs (~-aiSd) 1 + us (12.77), +aR .
it is seen that, assuming valid estimates of the field orientated currents, the speed can be estimated on the basis of vS q , whereas a change of the rotor resistance might be detected from v Sd . This is indicated at the lower end of the control section in Fig. 12.41. On the other hand, should a mechanical sensor be available for speed control, the structure in Fig. 12.41 becomes a normal field orientated control scheme, based on stator- instead of rotor-flux, where w' is replaced by the measured speed W. The flux acquisition is unaffected, still being based on measured voltages and currents. AIso, the outer layers of control are the sarne, whether the control is rotor flux- or stator flux orientated; they could again be extended to include a position controlloop. The advantages of stator flux coordinates are that the rotor resistance enters the flux model only indirectly, the stator temperature causing changes of the stator resistance can be measured and that the effect of saturation is already contained in the voltages. On the other hand, the stator flux model is more sensitive at low frequency to detuned parameters such as Rs compared with the rotor flux model shown in Fig. 12.16 b; also, measurements of the terminal voltages, the accuracy of which is quite criticaI at low speed, are needed as input and the magnitude of the magnetising current is produced by an integration with nonlinear feedback instead of a linear lag termo
12.5 Control of an Induction Motor Without a Mechanical Sensor
295
It is of interest to note that the current components iSd, iS q defined in stator flux coordinates are not completely decoupled, as was the case with rotor flux coordinates; there is a slight coupling effect similar to armature reaction with a De motor. This becomes apparent when transforming Eq. (12.68) into field coordinates; in steady state, Le. with constant values of ims,isd,isq,w,wms = WI, and W2 = WmS -w we find . ZSd
.
ZSq
. + W2 a T R ZSq
ims
= - -1 + as
T R (ims -
= W2
Solving for
1 +as
iSd, iS q
(12.78)
,
.) - a ZSd .
(12.79)
results in
A
':~
!"9__ ___ i sd~.~
r"·
!~~ --
_.._ ,
ims
o
ri s
..50
, _
_
__
_
'
J
(ORm
00,
- - --.,.
--
w 3. 0
2.0
1.0
t
mL !
ris
4.0
a ~ ~b Af o ~---,{ \. ll\ [h\ !X.\ A
100
k
\ '.1
.. 100
ims
Js jms
b
~
i ms
1..".
\ ,/. !;
\ ''/ /i
..
'I
\ \!y-.. ............................._...........-...........-.........I r I s
'~IL _.f-~,c~"...'i:.~~,L----_.
. \= .. . ..
o Fi~~ . 12.12. Sl.ItI,o j' nux ori (: lll.al.l' d ~ tll.l.()r CIlJ'rCIl!. v"dor
IUI I~ r'llld.IOII 01' .... r" ''1II1'''CY
. !f
~7
0.5 00.o
. ~'_ .
/ .0
__. •
ris
00 .
2.0
i
3 .0
4 .0
Il s
Fig. 12.4:J. Silllldlll.i oll res ults 01' tlu: 111.1\1.01' II" x Ilril:lIl.u.tcu induction m ot or control 1" l fl, . 12.-1 I, IIHill 1l; l/U'ntHlI·.,,1 /l 11I'. 'd f.,. :dIHlCk (n) Sl.n.r t.I "I'" rl·v,,,',,I,,!'. Itllfll",\di,,/ \ I.rll lll ll "'''/,II; (h) l" l\dillf~ I.r'lll/:;I'III.:: ai. ~. ,' I'I) :1 1'1 " '.1
Hch u lllI l Il hOWII iII
296
12.5 Control of an Induction Motor Without a Mechanical Sensor
12. Control of Induction Motor Drives
. tSd =
1 + (J" (W2 T R )2 ims 1 + (W2 (J" TR)2 1 + (J"S '
.
(1 -
tS
q
ims
W2TR
(J")
= 1 + (W2
(J"
(12.80)
(12.81)
1 + (J"S '
TR)2
where W2 T R is a measure of torque, as found in Eq.(12.3). The two functions are combined in Fig. 12.42; if the stator fiux is to be maintained at constant leveI, iSd must increase with the load. This could be achieved with a fiux controller. 12.5.3 Silllulation and Experilllental Results Simulation results with the scheme in Fig. 12.41 and based on parameters of a 22 kW motor are seen in Figs. 12.43 and 12.44; they show starting and A
reversing as well as loading transients at zero speed reference. ln Fig. 12.43 the "measured" speed w and in Fig. 12.44 the "estimated" speed w' serves as feedback signal. Naturally, there is little difference as long as the motor parameters are accurately known but, as mentioned before, the fiux mo deI is sensitive to detuned parameters at low speed. The only minor difference in the simulation results should not lead to overoptimistic conclusions.
ül
J
v_'~
1.0
0.5
3.U
2.5
II/
4.0
_,I,
4.
".'u
'-'N
ris
Esrimated speed (Feedback signal)
I I \.-
ISd '.' ....... )
50
\ 1\~YY\=f"'!.e 3.5
1.5
w'
.-
Measured speed
I 02 ülO
iSq
100
297
~_
I
ms I/
115 I \
-50
(
--
~
roo
ro,7
1.0
a A
I
mLi
3.0
2.0
4.0
h= , -'~I'
100~~i~~\ !xV'SLA'\ Vi V' . ..
100
A '00
lJ
V,- ,i
I
~
i
',./
'- ./
v
v
'-'
115
I
. ,
0.5
w'
03w 10.
o
~ tls 6
~ w'
,~
\
Ã- - -- - -- - -- - rv' , ,
7
8
9
v 10
·-Í ''
Estimated speed (Feedback signal)
L -"
Il0A
~
_
r~
_
o,
fi",
I" i ,:. 12.1\/(. Si ll' l'I lal.ioll "·'IIIII.H of" I.ltc Hl.al.oJ" .".i
1\ 1\J\f\ 1\ 1\
I
18 Nm
:=1
1_ _ _ • •
Statorcurrent
ç. 1\ I\~ ~ 1\ ",-=nrv v ITVVVIJ"'\::7""2/V V
wo ' '-- ' --~
4.0
~~ tis
E --~ ----t-+ '~'w
3.5
\]v" : . . . / .... ............... .
_
-- --- -
()
~
3.0
115
.'
'~
2.5
Fig. 12.45. Experimental results of a stator flux orientated induction motor control scheme similar to Fig. 12.41, using estimated speed feedback, transients following step changes of speed reference
CftJel [------
I
2.0
1.5
1.0
0.5
tis
Load torque
~-~ IIs .
Ji'iv;. '12.1\ (1. 1';)i[)('r i ""·1I1.1l1 reslllb: uI' 1\ ,, 1.,\1,,", 1111 X uri'·"(.I\.t.,·d illdlld,ion mot.or ('.()IILrol ~:<'I"."l<' lI i!!, i l "., 1.1) 1,', /,.. 11 .1 1, II Hilll, " '1 1. 1. ",,1. .,.1 ''1 ".,..( 1'(., ..11>" .. 1<, I.fJl.IIHi"IlI.H r()II()\Villl~ nle'" ( · r l ,t ~ III !.(· tl (Ir IqH eI l,tnqtl4 ' HJ, VI'I,' y IH IlV IJ I'( W .!
298
12. Control of Induction Motor Drives
12.6 Control of an Induction Motor Using a Combined Flux Model
299
Some experimental results with an "encoderless" induction motor drive have been obtained with a 1.5 kW standard motor, fed by a commercial IGBT inverter switched at 8.8 kHz and a control scheme similar to the one shown in Fig. 12.41 [J16]. Transients following step changes of the speed reference are seen in Fig. 12.45, with the measured speed also plotted for comparison. Fig. 12.46 shows the transients following step changes of the load torque corresponding to ± 0.4 of nominal torque at the very low speed of 0.03 w o , indicating a remarka.bly good performance under these difficult t est conditions.
are again the "voltages behind the transient stator impedance". Eqs. (12.82, 12.83) constitute the "volt age model" in rotor-fiux coordinates. The "current model" was defined by Eqs.(12.33, 12.34) as
12.6 Control of an Induction Motor U sing a Combined Flux Model
w' obtained from the combined evaluation algorithm. Of course, when both
TR
de
-d t
di
.
.
mR --;tt + ZmR = ZSd ,
(12.85)
ZSq
= WmR = W + TRZmR ' = W + W2 ,
(12.86)
where the measured speed W should now be substituted by a suitable estimate
From the preceding section it became apparent that employing stator volt ages as inputs to a "voltage model" based on the stator voltage equation, Eq.(10.50), has weaknesses in the low speed region because the terminal volt ages decrease with speed until they are eventually dominated by the voltage drops across the temperature dependent stator resistors and have little bear ing on the magnetic fields in the motor. AIso, use of stator fiux coordinates creates additional coupling terms between the stator current components in licld coordinates. With the "current mo deI" based on the rotor volt age equa tion, Eq.(10.51) and with rotor fiux orientation corresponding to Fig. 12.16 b, these coupling terms are not present and zero speed is within the validity rallge of the model; however, a speed signal is needed, which is undesirable ;1.11<1 should be made superfiuous. Hence one could consider a combined so IlItioll, where the two fiux modeIs in rotor fiux coordinates are computed ::i'JlltlLaucously and a speed- dependent selection of their results is made, :;lIch that the signals from the current model are preferred at low speed and I."OS(~ from the volt age model at higher speed. 'I'hc "voltage mo deI" is obtained by converting Eq.(12.51) into rotor field coordinates and splitting it in real and imaginary parts (for the real part this was alrcady done in Eq.(12 .52). This results in CSd
= (1 -
(7)
Ls
dimR
--;tt = USd -
. ' disd Rs ZSd - (7 Ls dt
+ WmR
(7
. Ls ZSq,
(12.82) all
(1
eSq
-
(7
. WmR = uSq - R S ZSq ' ) L S ZmR
(7
L S -dis-q - WmR
dt
(7
L S 1'.Sd,
(12,8:1)
models are simultaneously processing the measured currents and voltages, they are bound to produce somewhat different results for the field orientated quantities, indicated in the following by an additional subscript S and R. The control will employ a weighted mean of these values, depending on the estimated speed. There are countless possibilities to realise this by software; for instance, a speed dependent selection could be realised with an observer in the frequency-domain [J5, J6] processing AC signals in stator coordinates or, as is alternatively suggested here, an algebraic weighting function may be employed. An example of how the combined fiux estimation could be implemented is shown in Fig. 12.47. ln order to avoid a differentiation of the current signals in the voltage mo deI, an observer-like structure is chosen, where the signals esds, esqs are obtained implicitely by matching the predicted and measured current signals iss and is. This results in a lag effect that should be duplicated in the current model for obtaining consistent results. The combined fiux estimation vector containing i~R' w:nR' i' Sd and i' Sq which serves for the control of the drive may be defined as the weighted mean of the volt age and current model estimates. For the example of the angular velo city of the fiux vector this results in W~R
= f(w') w:nRR + (1 -
f(w')) w:nRS'
(12.87)
where O :::; f(w') :::; 1 is a speed dependent weighting functionj correspond ing definitions are used for alI other estimates derived by the fiux models. With a suitably shaped function f(w') as shown in Fig. 12.47 the control is purely "current model based" at zero speed and converges to a "voltage model based" operation at higher speeds, when the terminal volt ages are of sufficient magnitude and the uncertainty of the resistive voltage drop can be llcglected. As indicated in Fig. 12.47 the estimated speed signal w' to be used iII thc weightillg function f(w') as well as for the encoderless speed control lIIay he ohl.ailled hy combining Eqs ,(12.83, 12.86),
w!t(,I'I' !<:: (I)
('
i ii
I': ;,'
I
:i ( ':;'1
I
(IL. . ~ !J)
( li
(I
(/) I" , í,,,,u (1'::/1 , I
na i S 'J (I I
ti II )
). ~
(12.RR)
300
12. Control of Induction Motor Drives
12.6 ControI of an Induction Motor Using a Combined FIux ModeI
301
If the flux model is implemented in stator coordinates, processing variable frequency AC quantities, the speed dependent transition of the estimates from the current-based to the voltage-based section could be done dynamically by using the frequency response of the observerj again, at low speed the results of the current model would dominate whereas the signals obtained from the voltages would prevail at elevated speed.
&!
~I
c;)
&!
~ E
r ~ .S;; c:~ I l-E ~.~~ ~ I~~~ 8 ~c:>~~ :s. ::J
o:: o:: :-...5'
0lV)
CC C\j
>
C\,i
'+-.
-..;::
CC C\j
>
"
vi
&"
8 1 81
..... (l)
h~ 1
h
ê: (l)
ço cc ~ L0
cc
2a
C\j
c:
oS
.~
"Q
.g
(l)
«l
~Q ~
Q)
Q iJ '-
~.§
~
r.nVí (l)
~a
t t t
E: (l)
>
"
vi
&"
,
Q) "Q
a E:
......
Ol
c:
~
(l)
~
t::
::J
SJ
~
'
-o...
ti
::J
<'A
::J
.~
'
<'A
8
Meu ::;urenll !/Its
l" ifl:. I :lA7. (:"".1,;"",, V()Ii.:Ij·.' '/ cIII·,...·III. 111,,01,·' f"r r"L,,)' IIIIX ,·,;L;lIlilL;()II
13. Induction Motor Drive
with Reduced Speed Range
Some mechanical loads, such as fans or centrifugaI pumps, exhibit a strong dependence of the load torque on speed, so that a limited speed control range suffices for achieving the desired effect. The sarne is true for rotating converters employing a flywheel, Sect. 7.4j since the kinetic energy of the rotating masses varies with the square of the speed, there is little incentive in varying the speed by more than, say 20 or 25%. Similar situations exist if an electrical island grid with a fluctuating frequency, for instance a railway grid, is to be supplied from the public constant frequency line through a rotating converter, or conversely, when a variable speed generator is feeding power into the constant frequency grid, as may be the case with wind energy plants or pumped storage hydro setsj using variable speed there would improve the aerodynamic efficiency of the wind rotor with changing wind velo city or the hydraulic efficiency of a pumpjturbine with varying head. For applications of this kind a wound-rotor induction motor drive presents an interesting solutionj if the stator is connected to the constant frequency grid while the rotor is fed with slip frequency by a static converter, its rating is mainly determined by the desired speed range Llw and can be kept relatively small. Still, the slip power may be substantial with large machines, calling for an efficient control scheme. Beginning at about the tum of the century, a variety of drive circuits has been developed that made use of auxiliary machines such as rotary frequency changers [L23], they are of course obsolete todaYj with the progress of power electronics new solutions have become possible here too, one of which will be discussed in Sect. 13.1. The simplified mathematical mo deI of the symmetrical induction machine derived in Chapo 10 can be used for this purpose with minor modifications.
13.1 Doubly-fed Induction Machine with Constant Stator Frequency and Field-orientated Rotor Current Whell din ~ct <:111"1'('111: ll\ol,oJ') (.11(, :;!.iÜ,,!' I!lO!."!' 11 :111 111111': 1
is supplied to the sliprings of a wound-rotor induction C()ulwcl.(~d to the COllstant frequency line, the
"r which is
1./11'
Pl'lIpl'!'I,j( ' H III' il fI .Y11('!tr' olloIlH Hladtirl(~.
Constaul'. (.orqlw
304
13. Induction Motor Drive with Reduced Speed Range
13.1 Doubly-fed Induction Machine
can only be produced if the rotor is in synchronism with the stator field; the machine can also deliver reactive power, but at the sarne time the problems of Lhe line-fed synchronous machine become apparent, which includes starting :Iml synchronisation as we11 as oscillatory transients and pu11-out torque. This i" st,ill true, if instead of the direct current, a constant frequency alternating <"Ilrrent excitation is applied to the rotor winding; only the speed ofthe motor IIIIIS!. he different for the stator and rotor ampereturn waves to move again iii sYllchronism. For these reasons doubly-fed machines operated at constant rol.or frequency are not particularly attractive. 'I'hc situation is different, however, if the AC excitation of the rotor is IIlad(' sircd speed range. 'I'll(~ iilJpressed rotor currents are derived on the basis of the rotor angular i'",.;i Lioll 'l.Ild t.l)(~ vector of the line voltages; speed and reactive power control wli,1! ('III'('('IlL lilIlit are superimposed. It would also be possible to choose a III,d,or willl a Lwo phase rotor winding instead ofthe usual three phases; the "11<'1":t.I,ioll wOltld oe the same, except for the harmonics on the line side caused I,,Y III(' ,;wil.chillg of the converter [B27]. For the fo11owing considerations a 1.111"'1' plrasc rotor with N R = Ns is assumed, as in Sect. 10.1. '1'11<' lllat.lwllIatical model of the symmetrical AC machine, Eqs. (10.50 I(I,~):\), t.hat was derived with various simplifying assumptions can be em "Ioy(:d Itere also, it only needs to be adapted to the new constraints. The (oLor willding is now fed by a converter, ,
/l/(ZI!
I·
diR
L R dt
+ Lo
d (. - 'E) dt '!:.s e J = 'QR(t) ,
(13.1)
'/l.I!.(t)
Wound rotor induction motor
iS, us,
.~: 71Rl
+ 71R2
' 2
eJ'Y
+ UR3 eJ
_.
'Y
= UR12 - UR23 e J 'Y
(13.'2)
1:: Llre vector of tlw rotor voItages suppliecl by the converter to the sliprilll(H 01' IIH' 'llad,ÍiJ(', W1Wll ngain assulllillg thal. Ul(~ COllV(~rt;er is c<jllipped wit,h I"n nL !"'HIHII ", lilll', ("llnl'IIL ('oll(,J'ol J()()P~i, I.he' I'III.O!' (".111'('1'111.:; -i/{(t)
OJ,
e
a OJ
Operating range
FLlOJ
t Generator
b
Motor
mM
F ig. 13.1. Wound-rotor induction motor with variable frequency rotor supply (a) Circuit; (b) Operating range
handwidth, the rotor volt age equation Eq. (13.1) has no significance with re gard to the dynamics of the drive; this results in considerable simplification of the drive controI. The situation is in fact quite similar to the case of the stator-fed induction motor; an added simplification is that now both, stator ;llld rotor currents, can be readily measured. It is thus appropriate to fo11ow ;lg,ún the field orientated approach. ln analogy to Sec. 12.5 the stator volt age equation (10.50) is written in Llle form Rsis
wlr(')"(~
305
wl,,~r<~
+ Lo
:t
[(1
+ O's)is + iR ejE] = 'Qs,
(13.3)
acr.ording to Eq. (12.66)
i m ,o..;.(t)
= (l+O's)i-s +i-R ejE=i m s(t)eJI"(t)
(13.4)
\11 11.1.1 ('X t.(~Jl(kd lllagm:tising (:nrrellt. vcr lol' rf's ponsiblc for the stator flux 11I"ItIl!illl ': !(' a lla)!..' , I"i ~, 10,7 I' Sh olll,) 1.1,,' I:YIlIlIld.rical t.hree-phasc HlIPP!Y Ln wllil'll 1.11<' nl. ft.l",,' in "'01111,', ' 1,,'<1 , .. -dl lhll i i lI(d, jc"I1.I.I,' illl.e rll ll l i lll]I<'dil.II('(',
306
13. Induction Motor Drive with Reduced Speed Range
13.1 Doubly-fed Induction Machine
RN + jwoLN, for instance caused by a transformer or a section of the power line, it can be included in the effective stator transient impedance Rs
+ j Wo as Lo ==
Rso
i e ie
-R
307
Stator flux axis
(Um5
f·
1mS
+ RN + j Wo (aso Lo + LN) ,
(13.5)
with Rso + jwoasoLo being the transient impedance of the machine proper. Eliminating the stator current fs from Eqs. (13.3, 13.4) leads to
dfms Ts ----;{t
" + lms
_ 1 + as - ~:l!s
+ "IR ej
ê
(13.6)
.
Rotor axis
u-5 "'" e-5
Ts = Ls / Rs is the main time constant of the stator circuito The expression for the electrical torque is, Eq. (10.52), mM =
~ Lo 1m [fs ÜR e j ê)*] = ~
Stator axis
•
j
1 !oas 1m [fms (f R e ê)*] .
(13.7)
Fig. 13.2. Angular relationships of current vectors for doubly-fed induction rnachine
With the introduction of the rotor current vector in stator coordinates,
f R ej
ê =
iR(t)
e j (€ + ê)
,
(13.8)
the torque is mM
2
= -"3 (1 - a) L R ims iR
sin(Ç' + e - J.L)
2
= -"3 (1 - a) LR ims iRq . (13.9)
where Us is the RMS-value of the line-to-neutral voltages. Transforming Eq. (13.6) into field coordinates with the help of Eqs. (13.4, 13.8, 13.11-13.13) and splitting the result into real and imaginary compo nents yields two real differential equations for the magnitude and angle of the magnetising current vector,
dims. 1 + as Ts ----;{t + ~mS = ~ USd
.
+ ZRd ,
(13.14)
Thc position of the rotor current vector in field coordinates,
(13.10)
t,+e-J.L,
ti'
rorf'('sponds to the load angle. The angular relations of the various vectors \rc SIIOWIl in Fig. 13.2. The instantaneous angular velocities are
de dt
dE, dt- = W2, 'L'h!~
dJ.L
dt = Wms·
= w,
(13.11)
=iRd + jiRq
(13.12)
are defined as d - q-components of the rotor current vector in a stator flux ()['ieut.a.tcd reference frame or as direct and quadrature components of the rot.or current "in field coordinates"; they are DC signals in steady state. WIl( ~ !l assuming a symmetrical three-phase system of sinusoidalline volt-o HI-',C:; bavillg the COllstant frequency wo, the voltage vc-~d.or rotates 011 a circlllar pal.h,
'I/. ' :
(I )
' )
,
3V2
( 1:\1:1)
(13.15)
(13.16)
USd = -2- Us cos(wo t - J.L) ,
.
(13.17)
are the field orientated direct and quadrature components of the line voltages. The d - q-components of the rotor current vector can be derived from the trleasured rotor currents iRl, iR2, iR3 according to Eq. (13.12); the transition is again performed in two steps by first converting f R to an orthogonal two phase AC system which is subsequently transformed into field coordinates. With an isolated neutral of the rotor winding we have with 'Y = 2rr /3 :~n(t) = iu
ejÇ
.
.,' 1,/11
,
. ]
where
:~
:~ J" (/:: /.1 .,'" I
as
~USq +ZRq
uSq = -2- Us sm(wo t - J.L)
= iR cos 8 + j iR sin 8
ó
1 [1 +
= WmS = Tsims
3/2
rcal and imaginary parts of
iR ej(.~+ê-J.') = iR ej
dJ.L dt
= iRl +iR2 eh +iR3 ej2 'Y
./3 (. . ) . _.
:J
T
1.J7.~ -
'/.11.:1
'1,11..,
I .'I Zn.I, .
TI ... 1I 111,1I"qllO ,"I. 1.11I 1I: :,r"IIIIHI.i,," iul." li.·I,1 ,··,,,,,,Iillnl...:: .yi..!'):;
(13.18)
13.1 Doubly-fed Induction Machine
13. Induction Motor Drive with Reduced Speed Range
:{08
. !.R
ej
("-JL) _ .
-
ZR
(€+"-JL) _ .
ej
-
ZR
8
ej
= [iRa + j iRb][COS(t: - p,) + j sin(t: = iRa cos(t: - p,) - iRb sin(t: - p,) +j [iRa sin(t: - p,) + iRb
cos(t: -
p,)]
p,)] =
iRd
+ j iRq
;
(13.19) Llw angle
t: -
p, may be interpreted as the rotor position in field coordinates.
again be applied. ln this case it means that the coordinate transformation ej("-JL) inside the machine should be cancelled by an external transforma tion e-j("-JL); with the subsequent phase split a balanced three-phase system of current references is generated. This is seen in the lower left portion of Fig. 13.3. The angle of transformation, t: - p" is readily available from mea surements of rotor position as well as the stator and rotor currents. The transformation from field into rotor coordinates, in effect a modulation, is based on the following algorithm !RRef
Wound rotor induction motor I
Converter with current contrai i
(1+<3S) 3-{2 Us
R3
I
--~-------------
2 Rs
I
I
I
I
I
+ i R2 eh + iR3 ej2 ']Ref = [ Z' Ra + J,Z.Rb] Ref = ['ZRd + J,Z.Rq] Ref e - j ("-JL)
=
[iRl
(13.20)
,
which results in
iRb
IIi
Rq
I
, .~ ., - , _LI
iRai
cos(t: - p,)
iRaRef
=
iRbRef
= -iRdRef sin(t: -
iRdRef
+ iRqRef sin(t: - p,) , + iRqRef cos(t: - p,) .
p,)
The equivalent balanced three-phase system of current references is, Eq. (12.37) ,
e j«.~)
.
2 '
ZRIRef = "3 ZRaRef, .
~
1 .
ZR2Ref ,
ZR3Ref
iR~ Flux control/er
-----
Contrai in field coordinates
msRef
e .j(e'~)1 Torque contraI/e r
Speed contral/er
iRqRef
Fi~. l:~.:i.
309
is(t)
of Fig. 13.3, showing the interactions described by the preceding equét COIlsiderable similarity exists with the block diagram of the induction 11101.01', Fig. 12.12, except that there is an additional electrical input in the rOrJlI of the line voltage USo The angle wot - p, constitutes the advance of l.!Ip voltagc vector lL-s against the magnetising vector !mS' Since the stator rirclIiL is couflcctcd to a low impedance source, the angle wot - p, ~ ~ expc·, ricll('(':; ollly slllaU cltanges. As in the case of the stator-fecl induction lllot'lr 1'\1(' lll;q~llil.lId(~ of !.lte fillx cau only be altcrcd t:llrollgh t.!w lag 1'5 , Wlt C[(,iI.H
1 '
= -"3 ZRaRef
1
.
1
'
+ v'3 ZRbRef, -
(13.21)
v'3 ZRbRef·
Clear1y the decoupled control scheme offers direct access to the d - q-current components. There are various possibilities for designing the higher levei control as can be deduced from a discussion of the steady state characteristics of the drive. At constant load torque and speed the rotor angle is t: = wt and the flux vector rotates in synchronism with the line voltage vector, WmS = Wo. Anal ogous to Eq. (13.13) the vector of the stator currents is, ignoring harmonics,
Control of doubly-fed induction machine in stator fiux coordinates
'l'ile block diagram of the doubly-fed machine is contained in the upper
= -"3 ZRaRef
= 3 V2 2 Is eJ' Wo t ,
Is
=
Is
ej'Ps
,
(13.22)
where Is is a complex phasor. Similarly the rotor current vector in stator coordinates is
part.
i (t)e j ,,=3V2 I e j (w2+ w )t=3V2 I e jwot 2 -R 2 -R
Liolls.
!.lH' qlJ;ldrn,L1JJ'<' clJrJ'(~JlL
'i/(I{
1'[ !.II(' ('oflv,'rl.n wi t h ,'III. 1" '1\ 1, /)( , 111' <'''
i:-; availahk [01' rapidJy (;olll.rollilll< t.orqllC.
('111'('('111. COJlI. !'01 iII
wl',1\ 1I I'I', lip,Hdi' Iltr" 1,111', jd" I\,
II.i>p r o ld lllal.('d hy
IIr (' , 111(1,'11 1 i II
ii.
t.1m'('
pIl H H"
li('ld " " ol' .l l llíd,,·/I
,' iI,11
-R
,
IR = IR
ej'PR .
(13.23) Illscrting these definitions into Eq. (13.3) leads to an algebraic phasor equa I.ioll
Rs
~s -I-
.i Wo I,s Is + j
Wllich, wil.h L ..., I : \. ,'I.
(I I
(7,';)
Wo
Lo ln = U50
,
T'II! i~j illll s l.J'II,kd hy Uw ('qllivaklll. circlIil.
(13,24) iII
Fil',.
:no
13. Induction Motor Drive with Reduced Speed Range
J.s
RS
13.1 Doubly-fed Induction Machine
1R
(JS LO
Is ~ Iso
= Iso Uso
LO
IR ] = Iso [1 _ IR ei (Hé-/ll] [1- ImS ImS Rd IR eJ. ó] = Iso [ 1 -I --- - J. -IRq] [1 - --ImS ImS ImS
'Rd
'ms 2~--~--~r---.---.-
Supply
~ ,' ({JS
o -t-II
- lls"Us Ps
under excíted
I
o
"'í,~ . 1;1.5.
Mech. power
w<wo
-1
TIl(' st.ator resistance may be neglected with larger machines,
R s « wo Ls, ~j() Lh a t the phasor of the magnetising current is determined by the impressed sl.aLor voltage,
+ (JS ) -1S +1-R ~ ~
Uso
-~. j Wo Lo
(13.25)
' I'he lIo-loa.d stator current (IR = O) is
Uso ' so ~ j Wo Ls ' IU:ll co' I':q . ( 1:1./.;') ;L~S I\'II( :~; Lhe !"mlll
Mech. power
wo <w
Fig. 13.6. Power flow of doubly-fed induction motor drive below and above synchronism
Decoupled orthogonal control of stator current
= (1
) 1L I P
'41s
III.'>
11/IS
RI
-1r---+----r--t
nq
Supply
over excited
I
1so
(13.27)
This indicates that the decoupled control of the rotor current in d - q components corresponds directly to orthogonal control of the active and re active currents as is shown by the orthogonal grid in Fig. 13.5. Note that this operation in ali four quadrants of the current plane is not limited to constant speed but applies to the whole torque/speed area, Fig. 13.1 b, accessible with the voltage, Current and frequency limits of the converter. Depending on the operating condition of the drive, power is flowing from the rotor via the con verter to the line or vice versa as is shown in Fig. 13.6. The phase sequence of the rotor currents is inverted below and above synchronism.
Fig. 13.4. Steady state equivalent circuit of doubly- fed induction machine w ith impressed rotor currents
~
311
(13.2(i)
When assessing the reactive power balance, the reactive power drawn by the converter from the line must also be taken into account. If a cycloconverter is used, its lagging reactive input current may be substantial; it could impair the ability of the drive to operate with overallleading power factor. Measures for reducing the lagging reactive line current of the converter such as the reduction of its supply volt age with a transformer when operating the drive a t low slip are frequently applied. Starting the drive is accomplished with the help of resistors, as seen in Fig. 13.1 a; as soon as the motor has reached the operating speed range the current-controlied converter is switched in. The orthogonal grid in Fig. 13.5 provides a clear indication of how the higher leveI control should be arranged. As the stator flux (ims) is essen I.ially prescri bed by the line volt age Us, the d-component of the rotor current wlII
,·olll.ml LUI
312
13.1 Doubly-fed Induction Machine
13. Induction Motor Drive with Reduced Speed Range
I.
-I
Microcomputer 8086
I
's react
iSI iRI E I I ,-------''---,
(ú
In1
iR1Rel
100
Us
.---::,------, is I
Active
Voltage feed-forward Off t. On .
iRIA
!,.Js I
's act
313
50
iS
o I \\
~
H
II
H
H
~I I
I
\
, II
I
Vs
\
ReI
úl
o Reactive current control/er
Field coordinates
: Coordinate , :transformation:
0,25
0,5
0,75
Fig. 13.8. Effect of voltage feed-forward on current control with a cycloconverter converter
Rotor coordinates
(J T R
Fig. 13.7. Contrai scheme of doubly-fed induction motor drive with microcomputer
. di R dt + ~R =
avoid undesirable operating conditions and to protect the equipment. The cOlltroI section in the Iower part of Fig. 13.3 represents only one of severaI [JossibiIities; yielding torque/ speed curves with torque limit as in Fig. 13.1 b. Figure 13.7 depicts the controI scheme of an experimental 22 kW drive wi[,1J microprocessor controI [AI3]. The rotor winding is supplied from a 6 Pllbe cycloconverter with analogue current controI; the converter was chosen 1.0 Ilave adequate power rating up to 7.5 Hz corresponding to 15% sIip. To illlprovc the steady state accuracy of the current controlloops, feed-forward voll.age :;ignals were added for counteracting the slip-proportionaI induced voll.agcs in the rotor windings. As was mentioned before, the angIe f.J, of the magnetising current (stator flll x) UUl either be computed from measurements of the currents, Eq.(13.4), or fr()J\J stator voItages and currents; in combination with a measurement of the ]"o[,or position ê this provides the information for the modulation producing 1.11(' reference signaIs for the rotor currents. Wit.1J field orientated controI an excellent dynamic performance of the dri ve is achieved, even though this may not be essentiaI for some of the ap plications rnentioned, but the transparent structure of the controI is beneficial a~; i t. ofrcrs a great degree of flexibiIity. I<'igll[(: 13.8 shows the effect of voItage feed-forward on the performaucc 01'" I\. ("II1TI'III. loop ill stl'ady.,state. The cyclocollvertcr is operatillg at a freqll(~nl'y clr (i.'{ 11'1. c'C))TI':q,()l\dilll~ to a sp'~I:d of 1.700 l / ll1ill 01' I.h(~ I\ -polf: llIH.r.hiw: . Tlle c!loi .. c' 01' I.!II' II'c'eI rorwllrd Sr.lII'IIW i s lJ.tSI'd OH I.IH'. l'Ot.or v()lI.n,~I: cqllat.iolJ I';q ,
wil,h
(
1 - (J
)T
R
d (' -J' E) dt ~mS e .
(13.28)
The Iast term may be expanded in the form ~R =
(1:1 I) wllll'!I, i 1I1.l'ocl IlC' i 1'1', 1.llI' JlHI,,:lJdi :II! 'I'. "111 1'('11 1" ill
1
RR Y.R -
[,Ii/ /(11
'1'/1
d .
(1 - (J) L R dt (i.mS e- J E)
= (1 - (J) LR
= (1 - (J) LR
:t
(imS ei (J.I-E))
[d~~s + j
(13.29)
(wmS - w) i m
s]
ei (J.I-ê)) ;
with ims :::::; const., this represents the sIip-proportionaI voltage acting as a Ioad disturbance in the current controIIoop; its effect is compensated by feed forward with the necessary signaIs being generated in the microprocessor, Fig. 14.17.
Ps Os Pr' asO
PSRef
0.8 0.5 ,"' ...
'a
'"
\.
. _ . . . . . . . • '... -"
ú)
= const
tI
o I _ .. • ov;:/""9' o 100
·(.....,·- .. I..tI--1P....í>·V·=:ff' *{P
200
300
* ..... '"""1' 400
..
tlms
FiK. 1:1.11. Ik,:cHlpl, ·d rnntrol of Il.ctive/rcn.çtivc powcr with a cycloconverter
13.1 Doubly-fed Induction Machine
13. Induction Motor Drive with Reduced Speed Range
:H4
The measured step response of the control loop for active stator power (with the speed control disabled) is seen in Fig. 13.9j it exhibits good dynamic pcrformance and little cross coupling with the reactive power control loop. ' l'he ripple on the recorded traces is caused by harmonics of the cycloconverter a.lI(! quantisation effects.
30 V' 16 A OJ
O
1
OJ
istics of the drive, which may also be employed for power factor correction, apart from high dynamic performance drive control [A31, H36, H37, K17]. Even though this type of drive shows very interesting characteristics, the amount of hardware for the rotor si de converter is considerable which is likely to limit the use to higher power and special applications, where the added flexibility provided by two-axes control is important. One prominent area of application is that of generators in large wind power stations operating on the constant frequency gridj the variable speed operation uses the high-inertia wind rotor effectively as a flywheel in gusty wind conditions. This helps to smooth the power fluctuations and reduces the rating of the generator as well as stresses on mechanical components such as shaft and gears. Another ex ample are constant frequency ship- board generators, driven at variable speed by the main Diesel engine with its low fuel cost, so called shaft generators.
10ms 1
30 V 115 A
p-ê
Contrai
p-ê
w,
URq
O URd
30 V
I
OJ
315
Variable speed generator
16 A
O
, I"i r.:; . I :J.10. Waveforms of totalline-side current of a 22 kW doubly-fed drive with ::Y "IIII<'I.rical GTO voltage source converter at different reactive power setpoints. I.ill <: C1ment (a) in phase, (b) lagging, (c) leading
i ,
:_--(~~;-.--+Load impedance j
Driving forque
p
(Mo dei of AG grid) I
OlmR
Whcn employing a De link PWM-converter instead of a cycloconverter for sllpplying the rotor-winding, the performance of the drive can be further illlproved in view of the higher switching frequency and the reduced harmonk illt.(~raction s on the line- and motor-side. The reactive power drawn by tlw collverl.er from the line can also be greatly reduced. Fig. 13.10 displays the to t.al Iill!: currcnt, comprising the stator as well as the converter input currents, WllClI a sYlfunetrical volt age source converter with GTO thyristors, as showl1 iII ICi g. 11. I :3, is used. By designing a decollpled orthogonal control stnl(:tlln~ ;d li O for (.jl<~ líHe sid e converter, ha:wr! OH tJl<~ v()lt,:q~( ~ vector , th(~ active fUld l"C °ndi VI' p:....'t.K 01' t.Il( ~ t.ot.nlliJl( ~ Cllrr(~I1t. c;t.u h(~ eOlltl'oll( ~ d, as di :~ctH;sed il1 Soei.. r:! :.!.j I.hi !1 1'I'lJvid.,:; ('o"lpld(' [n '('d""1 wit\h 1 ' (' I ~n, l'd t.o 1.111' liIHHlíd., chnrad .•'r.
~'-------------------------------------------' --+-p
S/a/or coordina/8S
Field coordina/8s
Fig. 13.11. Block diagram of a variable speed AC generator with impressed rotor voltages feeding an island grid at constant frequency
If the doubly-fed machine is normally acting as a generator, is is prefer able to operate the rotor side converter as a variable frequency volt age source, i.(·. wit.ll
316
13. Induction Motor Drive with Reduced Speed Range
13.2 Control of a Line-side Voltage Source Converter as a Reactive Power Compensator
f~' fI ~ ~~ ~==:;; 1,5
0,5
~ US~
, fV
0,5
2
~
2,5
tis
2,5
tis
The principIe of field orientated control as discussed in Chapo 12 with the example of machine-side volt age source converters can equally be applied to line-side converters for achieving high dynamic performance. This is shown in Fig. 13.13 a, where a symmetrical voltage source PWM converter serves to connect the DC link of a doubly-fed machine through a transformer to the constant frequency grid. Line-side PWM converters are applicable whenever a DC bus is to be coupled with the AC grid, such as fuel cells or variable
~ 1,5
2
URA ' URB
r··~ 0,5
1
1,5
2
2,5
1
1,5
2
2,5
l
y
"mM,O)
«Moto~\\::::~::
s=p+jQ N N N
,,'.,~~\\ Gen..:/",1.'..
.
'OM
-,' '-: ',---.'
"
"
ma;~ine sid~1onverter
iON / ': r-~__1,..~r_~__\----
tis
•
J
F ig. 13.12. Simulation results of a variable speed AC generator with impressed "d,!lr voltages. Transients following step changes of the volt age reference and a ramp (:IJa.lIge of speed
coordinates; the stator- fed AC grid is represented by a load impedance /I /" /, f , in stator- and the electromagnetic interactions in rotor flux- coordi II;t.I,C S. The structure of the diagram is similar to that shown in Fig. 12.38. f\ I.w,,- axes control is indicated for the stator voltage and stator frequency; will, the help of two separate PI- controllers, the simulated results in Fig. I:\. 1'2. are obtained, showing the transients after step changes of the voltage l'(,r(~l"ence with regard to frequency and amplitude and when the mechanical spccd follows a ramp change; the sequence of the rotor voltages reverses as I.ltc rotor passes through the synchronous speed. Future high power applications for this type of control are found in IHllllpcd storage hydro power stations for changing the speed of the tur bilw/ pump when it operates with varying head, thus improving the hydraulic dli(:icm:y. AIso, changing the speed of the motor-generator-set by small alllollllts allows to activate kinetic energy and thus offers the option of COI1 I.lilHll,illg to thc dynamic stability of the powcr grid [H36, H52, H75,I<17, T3i .
. . . . . .
.
Uo
a
_II':('.
,
I
I I
.,;: : lt' <- .... '
•
1
,
I
I
I
II) :(
I
....- - -......
,
I
I ' I
I
,
l(
-,~;
I-r'
I
,
I
I
I
I I
•
I
II)' :( :{:
-:-' I
-r'
_II' I
I I
I I ,
--o-.. --.--------~---~-------- .. ---'
line side converter
solar converter
.. ~·o +.-;~".. ..•-['::::":0 -. ................ _.,. .. ,
Solar radiation
I '
\.
:: : :
'OS
'S
::
i S
/ '
'
: I
U
MPPl
O
:i'; .,:,-
_II", '-. . '
"
I
--!---
S
- - - - -O -. -
b
'"
I
~1.~ ~J~ ~;J ~;J ."'':'"
, I
I
I
-,,-
I
:""i: I I
:~ :~
•• I.
.,-
" ,.
-'f-
, ,
--
~W: :~: ~: :'~: •••
,
Us
'
~~.I~ ~ :..t~ ~ :..~)~ ~..y '
: :: :: :: : ~-~ ~ . ~ ~ . ~ ~-~ :Jr. :Jr. ~ ~~,~ .,
U
:
c
r,
+---+--_.
:/~,-
I I
t.
I I
,
~--~---
:(
-J:
_II~
I. ' .
_II' I
-1-\
,
I I
II):
:/~:
I I
"
_II'.
t--~---'
_II' I
L-...- - -.....>-....- - -...-
I
I I
I I
J
,I.( ,I"
,
• , )
1'(
~
~~
YN=Yo
tis
~~.-: 0,5
Supply
.....- - - - - - - _ .......................
---
uRQ
U RD
317
13.2 Line-side voltage source converter
•
" '
I,
j
I
,
••
-1--i--------~- -1..---~ [- ~ l~ -~r- ~ I
Step-up chopper
Solar panei
:Fig. 13.13. Line-side voltage source IGBT converter (II) in rol.m circllil. 01' a doubly-fcd AC motor, (b) for solar generatorj (c) "MHXilLIIIIII I'ow(~r PoillV' çonLro\ 01' ~()Iu. r pnlle1
318
13. Induction Motor Drive with Reduced Speed Range
13.2 Line-side voltage source converter
319
~peed
wind energy plants employing a DC Iinki a solar generator is indicated in Fig. 13.13 b, where a step-up chopper raises the voltage Us of the solar paneIs at the irradiation-dependent maximum power point to the leveI of the Iink voltage UD.
R o---CJ
L íNl
- ' íN2 1
~_.
iN3
U N1
--
j
U N2
J
Il
.
v v -I 23
-I 12
1 51
I
iOM
'_f2 r L
\
/
~c~r~
c, -,- I Uo
/ \ /. '- / 'Nd
úJo . A...... /
l·6
d/\ v ~ \ q I',
h .
/
fine currents
Q
\
/
ic
\V
!N(t) /.-.\. / I \ 'Nq
' ..... dÓ
6
h
... \
,6
1
,, \
j
t.Nb
LdiN dt
, \
, J R'LN
ó
,ó
'---- 53
--------------V --V--1 ---------V 1 2
LJitJ
T
-1
t
U N3
iON
Vector o, fine voltages
t i í l-I I
\
'! (t) .
3
I I
Vector o, voltages at converter terminaIs
jiNb
_'''!3 -.J
UNa
Fig. 13.14. Equivalent switching circuit of the line-side converter
An equivalent switching circuit of the line-side converter is seen in Fig. 13.14 corresponding to the circuit in Fig. 11.7 ai the transformer is rep resented by its line impedance. The time-dependent vector of the sinusoidal and symmetricalline voltages with constant frequency is with wNt = wot = ~ and , = 27r/3
Y,N(t) = UNi +eh UN2 +e j2 'UN3
= 3../2/2 UNei>\(t) = UNa(t) + jUNb(t)i (13.30)
similar definitions hold for the PWM volt ages at the secondary of the trans former, i.e. at the input of the converter bridge,
:11.(t) = Vi
+ eh v2 + e j2'v3 = V12
- V23 e-h = va(t)
+ jVb(t)
(13.31)
and the line-side currents
iN(t) = iNl
+ eh iN2 + e j2'iN3
= iNa(t)
+ jiNb(t).
1\
II"
I' I'I,,,
' NU)',· ,\
·~N.t I
111V'1
may be called line-orientated variables. The differentiaI equation for the line-side impedance R, L
L
diN
.
dt + R '!,.N = Y,N
-
(13.34)
:11.,
where R < < W N L holds for higher power, is transformed into the moving coordinate system by multiplication with e- j >.. and subsequently split into real and imaginary parts,
diNd . In/ 2UN-Vd+Wo L 2Nq, ' L--+R2Nd=3v2 dt
(13.35)
(13.32)
Control of the currents is again carried out in a moving frame of coordinates so that the feedback signaIs are DC quantities in steady state. According to Sect. 10.3 a suitable reference frame are now line voltage coordinates, wherc tlw vector Y,N assumes the role of the induced motor voltages ~s anel the i>uls( ~ width-lIlodnlat.cd converter voltagcs y(t) serve a!i Lhe aetnaLiIlf( variabJes. Fig. 13. Ui slIowH Lhe v{:cLor diag-ram of tlw lill(·~ · sid( ~ C(>lIv( ~ rt.(~ r. 'l'll(~ t.rallsforllwd voll.;I./',"" ; lIId '·III.,.c:"I.~l .,,(1) ,.
Fig. 13.15. Vector diagram of the line-side converter
( 1:1.:\:1)
LdiNq -dt
' + R 2Nq
= -vq -WoL'2Nd,
(13.36)
The instantaneous direct and quadrature components ofthe currents, iNd and i Nc11 are a measure for the active and reactive currents flowing from the line to the converter, as expressed by thc instantaneous complex power, 11."rv
'lN
(:IJ:~/2) lI N " .i>-' i N - :IJ:~/'2 (/NI'iNdU)
I " !i 1.( ·ll.dy nU d,<'
I jiN,,(l)Jj
(1:1.~7)
320
13,2 Line-side voltage source converter
13. Induction Motor Drive with Reduced Speed Range
321
y (v+ 1) e- jWot
,
I +
iNq
)
n'
iNb
iNd
PWM sector iNq
i Nq' control/er UD
.Y Ref
I~
I~I~
e-j wot
Ir:: ~l'
t
Nd
U
y(v-l) e- jwot
I_
ó+~-:~_----'-_ __
1m
iNdR • f
i Nd'control/er
Fig . 13.16. Volt age vectors of converter in line-voltage coordinates
I
Fig. 13.17. BIock diagram of the orthogonal controI of the line-side converter
t
l/lN
(iNd
+ jiNq) dt = 3.J2/2 (INact + jINreacd
(13.38)
t-TN
holds. For the purpose offast control, the instantaneous values iNd(t), iNq (t) llIay be used as feedback signals in place of active and reactive currents, "VI'H though they are only identical as mean values and with high switching fJ'c<[llcncy, i.e. low harmonic contento The converter acts as a reactive com 1)( ' llsator by producing a controllable reactive line current, whereas the active I iII\: cmrent has the task of controlling the DC link volt age UD. The mesh equation for the DC link volt age is, combining the machine-side ;\.lId the line-side link currents iDM, iDN,
C
dUD. . = tDN - tDM; dt
~-
(13.39)
·i /) M is a disturbance in the voltage controlloop. When the converter is used (!xdusivelyas a reactive compensator, i.e. without a machine-side converter, I.be constraint is iDM = O. The available voltage vectors in moving coordinates at the terminaIs of 1.11(' line-side converter are indicated in Fig. 13.16, from which the PWM collvcrlcr selects by vectorial modulation in each intervall'o the two neigh~ "O\lriIl~\ vec.tors. Th e switching fllIlctions Si == ± 1 correspond again to ti\(' !'ol.(:II(.ial nf (;11(: (".()\lvcr!;( !r lel"llliual i, as Hl1oWI\ ill Fig. 11.R; t:lw Jll a gnitlldc 01' 1,111' voll,iI,/',(' V('('!.!)!":; i:; dd.< !rlllill('d hy LI\(' 1>(; lilll( VO]l. i\l!;<',
= V12 -
= uD/2[(Sl -
S2) - (S2 - S3)e- h ] = uD/4[(2S1 - S2 - S3) + jV3(S2 - S3)] = UD e jvrr / 3, (13.40)
1L = v a + j vb
v23 e-h
where 1/ = 0,1,2...5. The switching combinations result in a six-pointed star for the voltage vector 1L( t); in addition there are the two zero vectors (SI, S2, S3) = (-1, -1, -1) and (1,1,1), when the converter terminaIs are short circuited at the upper or lower DC bus. When neglecting the protective intervals and applying the equivalences given in Eqs. 13.21, the link current iDN may be expressed by the line cur rents,
iDN
= SliNl + S2iN2 + S3iN3 == 1/3 [(2S 1 - S2 - S3)iNa + V3 (S2 - S3)iNb];
(13.41 )
during the zero vector intervals the line currents are bypassing the DC-link, iDN = O. The six-pointed star of the available converter volt ages in line-voltage coordinates, Fig. 13.16,
1L(t) e- j ).. = Vd
+ jV q = UD e j (vrr /3-w o t)
(13.42)
rotates with the angular velo city W o in c10ckwise direction and it is the task of thc current eontrollers to generate the needed direct and quadrature volt ;lj;(! COlllpüUClltS ('li,l + jVq)Rc! for tran5formation into the AC components ('(I" -1- J VI1)lkj'; til(! v(~dorial Illodlllat,or 1,11(:1\ dd
13.3 Wound-Rotor Induction with Slip-Power Recovery
13. Induction Motor Drive with Reduced Speed Range
:~22
is mainly performed in the vicinity of the real axis, using the two adjacent
available voltage vectors. Vd(t) and vq(t) are in steady state DC voltages wit.h PWM-related noise; because of the transformation, they contain no line I'requency components. The block diagram of the orthogonal control is drawn in Fig. 13.17 show iug in the upper left comer a demodulator for identifying the line vector from I.wo of the line volt ages and the usual transformations. The transformed cur r< ~IltS, representing DC signals in steady state, are controlled in two channels, with the undesired interactions removed again by feed-forward at the output ()f the current controllers. Experimental results obtained with a 20 kVA IGBT reactive compensator are depicted in Fig. 13.18 employing vectorial PWM with 4 kHz clock fre qllency [H36). The line currents shown in Fig. 13.18 a are nearly sinusoidal aud the transition from leading to lagging reactive current takes only a few ms. Changing the link volt age reference UDRef in Fig. 13.18 b results in a hrief active current component for charging the link capacitor, while the re active current remains nearly undisturbed; clearly the decoupling control is dfective. De link voltage
Reactive companent af fine current
40
350 >
Ir (j e·/
INq/A
o)
uD
/ 1/
300
I
•t 20ms _ _ _o
I~ I~IT
()
.
.
1
.
50
I.
~-a
The doubly-fed machine described in the Sect. 13.1 can operate as a motor or generator in a speed band above or below synchronism, drawing leading or lagging stator current from the line. Depending on the operating condi tion power :fI.ows from the converter to the rotor or in the reverse direction. This high degree of :fI.exibility calls for a relatively costly converter but sim plifications are possible when the specifications are relaxed. For example, when rectifying the three-phase rotor currents with diodes and feeding a DC link, the power flow becomes unidirectional. This is so because both volt age and current at the DC side of the rectifier are restricted to positive valuesj seen from the AC side the rectifier appears like a nonlinear, purely resistive network. The purpose of this scheme may therefore be to recover the slip power that would otherwise be lost in external rotor resistorsj it can either be converted to mechanical power in a DC motor coupled to the shaft of the induction motor (Kramer-drive) or fed back to the AC line with the help of a line-commutated or other type of inverter (static Scherbius-drive) as is explained in Fig. 13.19 a.
20ms
-I....l-i W-\ ff-
i,
JActl e il A
o
ilA
o
i
..,.,
(j)
1\ iNd
Reac ive iNq
\ '\,:;
I
ÜDmax
~
.
IS' - /' )
om, one ts af fine urre t
(j)
o - Ll
UR=O
---.L
~--
UDmax
·50
·30
Loadcurve
b
Fi~ .
13.18. Measured step responses on a line-side IGBT converter
wil.h v,~ct.orial PWM at 4 kHz clock frequency
(a) Itc·.versa.1 o[ Lhe rea.ctive current, (b) Change ofDe link voltage
TIl<' f')(;l.lll[lI,~ l1\akf~S iI. (~vid(:lI1. Uml. pow(:r .. kd.rol\ics, ()rl.( ~ \I look('d aI. a H I.l'ollld " H(l JlI(' lill(,- lii ,k illl.f'l'af'i.ioll~; iII !.lI(' fU)'I1I "f 'r, ·;"f'I.iv(' ('111'1"':111.1' 1.11<1 h ll,I""IO'f1 í( "II, (" :UI Itl nll provi,.!,' II. :;,dlll.illll rol' \'("'I<'Villl', :·"1<'1, ..t["'(,!.I'i; I.lIÍ II
II. ' ·' bll! if' 01'
13.3 Wound-Rotor Induction Motor with Slip-Power Recovery
30
j
; I.',()
is due to the emergence of semiconductor devices with adequate switching frequency. The cost and complexity associated with the line-side converter will in future be tolerable even with lower power equipment as industrial drives. At higher power the :fI.exibility will be important for the stable and economical operation of the supply grid.
Scherbius
-i
"1\ 1: lL
/lIV
4
, iN t
U23
1.'11)
~
/
uD" fi V
urfV
o . 'Ia
. ') /'
323
a
b
mM
Fig. 13.19. Wound-rotor motor with slip power recovery scheme. (a) Circuit; (b) Operating range
Ag;ülI, ~;[ip POW(~l" rc:c()very is of part.icular in(;erest at high power such as for )'ul.;~l.illl·. COllVf:I·l"rs with H.Yw1lf'd :iI. o)'al~ \' iu rolling llIill driv('!; UI' for 1,, ') 1"1 "",.,, plllll p l~ n!.f'lIdilll'; iII!." 1.11<' :~I) MW 1'11 1'11,<', ~; ('(.1.. 7.1\ , Whii<- iiI<'
I\( ~ (!ded
:~24
13. Induction Motor Drive with Reduced Speed Range
Kramer-drive requiring an additional DC machine is no longer of interest, the Scherbius-drive with the associated solid state equipment is still used today. ln view of the diode rectifiers in the rotor circuit the operating range is IlOW restricted to areas that would be available with external rotor resistors, i.e. motor operation below and regeneration above the torquejspeed curve valid for short circuited rotor, UR = 0, as shown in Fig. 13.19 b. AIso, since t.he rotor currents cannot be impressed at will, the stator currents are always lagging the volt ages as on a wound-rotor motor with secondary resistors. This cxcludes the control of the stator volt age which was possible with the more general scheme in Fig. 13.1 a. Speed control is achieved by changing the back voltage UD in the DC link. If UD is increased, the speed must drop so that the rectified rotor volt age can maintain a current through the DC link; without current in the rotor windings the rotor produces no torque. Closed loop control is usual with this type of drive in to have access to torque and speed and to protect the equipment against overload. Analytical modelling of the drive is made complicated by the rectifier generated harmonics in the rotor windings which also cause distortion of the stator currents. The subsequent discussion is only an approximation, a more cletailed analysis would have to be based on a digital simulation.
13.3 Wound-Rotor Induction with Slip-Power Recovery
W
I üOmax
Wo + L1w Wo
Wo - L1w
/UR=O
/
\ -
w = const
I~ fo=iomax
uOmax
mM Fig. 13.21. Torquejspeed curves of controlled Scherbius-drive
The DC link contains a line-commutated inverter for feeding the slip power back to the grid and a smoothing choke to produce continuous current over most of the load range. A step-down transformer matches the supply voltage to the volt age in the DC link, which is quite low in the usual speed range, and reduces the reactive current at the line side of the inverter which is caused by the delayed firing around a ~ i. Often the transformer ratio is changed in several steps to create optimal conditions over a wider speed range. The rotor current and, hence, the current in the DC link are roughly pro portional to torquej the inner current controlloop, seen in Fig. 13.20, provides therefore a good substitute for torque control and torque limit, while the speed controller is superimposed, generating the necessary current reference. Further control functions may be added if the need arises; for example on a fl.ywheel drive and with pulsating load, approximately constant line power could be specified resulting in a fl.uctuating speed leveI.
mL
AG motor with
mo úJ
iO
iO Ref
("1101
WRef
Speed iO'Rel Gurrent contro/ler controfler
Firing
circuit
"'ir;. 13.20. Control scheme of static Scherbius-drive
'l 'Ii,: circllit. of t.hc Sclterbius-drive :md t.hp lll-illal couLro[ schcllu: ClIlp[Oyilll, ali lIUII'1' ('uITCIIL- :1,11(\ ;Ul ()llt.(~r spec(l- 100p in S!JOWII iu I"il':. I:t20 . T he s t n ('t.illl ~ IT, N i~ I t. (l 1' Ih 'l'V(,:: 1.0 hl'illJ!; !,III! lIIot.or inl.n 1,111' 1;IW('ifl( 'd :;p( '('d I'ItIIl';(', W!II"'( ' 1,11(' 1'111.111' VIII "III'," Im H Ii llll ki "lIl.1y d"(' I'('IIIIe',d 1,/1 rdllllV ! l w l l.d d llJ ~ iII 1.1,,' (' ol1vnl., 'I'.
325
IãO
.!2... ioo
r------------------- I
converter in rator circuit
----1---: ~
Wo ~
Wo
Gurrent contral/er
Converter
i .
1
"!!.a _ U
-
RO
- S
-'
}.Q.. == mM '00
mo
•
!
'------------------------------ ~ Fil-':_ J :1.:.!2. ~~ i ll1J,IiIl"d I""cle diaW'll ll1
"r ScI\( 'rhiIlH-driv(,
13.3 Wound-Rotor Induction with Slip-Power Recovery
13. Induction Motor Drive with Reduced Speed Range
326 iR1 IA
io
50
A
o I "ri
'"
)
50
.1", 0,15
_,
i 1A . S1
327
0,2
tis
iORef
~=-O,17
'.V'
1
~-.A
• lO
S = - 0,17
25
A f\ f\ f\Yr6 !\ f\ f\ ( a ~5lJV\}\]\Jo·lJ V'J V VJ.2 ;.
Generator
O
iR1 1A
o o
25
50
75
100
ms
Fig. 13.24. Measured step response of link current control loop
50
O
II
\,
J
I
\
')
(\
\
J
iR A
I
'" ,<~1~\JVVVVV\J /\ D[\ f\ f\ fi ttM
50
;.
-50
O
s
b Fig. 13.23. Stator and rotor currents of Scherbius-drive in steady state operation ahove synchronous speed (n) S = -0.17; (b) S = -0.30
The steady state torque/speed curves of the static Scherbius-drive are secn in Fig. 13.21 for WRef = consto As long as the direct current is below the ClIrrent limit, the speed is kept constant by the PI-controllerj with increased load the speed controller eventually saturates causing the curve to bend down IIlltil the maximum volt age of the inverter is reached. Further drop in speed J11llst be avoided since it would inevitably lead to a commutation failure of t!te inverter as explained in Fig. 8.17. A simplified block diagram of the controlled drive is drawn in Fig. 13.22 wilh the control loops arranged according to Fig. 13.20j the normalising quan tities "no, mo, iDO etc. are suitably chosen. Thc inverter with the mean direct voltage UD is represented by a control ddny a~i dcsnilwd iII Fig. 8.31; UD is the badc volta{~e opposiug the rectif! (!d rotor volt.aLÇ(', wit.h t.h(' ditft:f(~ncc driving th(~ Cllrn~llt throngh Lh(~ De lilll< I~Jld 1.i11' rnl.m wiJldiH \{S . T" iH t.lte iilt.(~r t.illll: COl.lHl.aul. whidí iH IllUillly
~
= - 0,17
Generator
A
44
- 44
Fig. 13.25. Measured stator and rotor currents during step change of link current reference
As mentioned before, the details are of considerable complexity, in par ticular the current waveforms in the rotor and stator at different speed and loadj this is shown by simulated results and measurements conducted on a 22 kW cxp(~rilllclltal Scherbius-drive [A13]. Measured waveforms ofthe stator llnd ro!.()r CIIITI'IIt.S ai, two operating [>oillt.s in the generating region (above Hylldmlllull!l :;IIO','d) arl~ s(~( ~ n in Fi g. I :t2:\ wi!.h thc sanw din~ct Iink cnr rell!., -7 1 , rio f\, \11 Ilot.h ('asl':;; 1.111' 1-1 11('('d dl 'I" ' lIdl'llI, diri!,orl.ioll of i.Il(~ ntnl.())'
:l28
13. Induction Motor Drive with Reduced Speed Range
14. Variable Frequency Synchronous Motor Drives
US1
,./
s
s
~'t
1.1
1.2
1.3
1.4
1.5
~I ~ ~ ~ fl ~ ~
1.1
1.2
'p I_
:
$:
i,!
1.6
S
I 1.6
S
I
~
~ • ~ , ~ ~~~~~~~ A
1.3
1.4
1.5
I;'ig. 13.26. Simulation results with a Scherbius drive
currents is c1early visible. A recorded step response of the inner current con trol loop is depicted in Fig. 13.24 showing good dynamic performance of the dectronic torque controI. The effect on the stator and rotor currents when changing of the link current between O and 50 A at constant speed (1725 l/min) is shown in Fig. 13.25. As an example of the results obtainable with digital simulation, Fig. 13.26 di spl ays some computed curves of voltages and currents as weII as electrical I.orque. Clearly, the deviations from sinusoidal waveforms are substantial; of particular interest are the long commutation intervals of the uncontrolled n~ctifi(!r, caused by thc small slip-proportiona! driving vo!tagc and the ro(.(11' lpaka/'.(! n'iLct rmce. On the other hand, rcclncing thc higher freqllcncy hanuOll" ic: ; (,(~lIds
(,0
:aJlool.h tll( ~
CUUC1ÜH alI(!
I.he !:orqll(!.
The speed of a synchronous motor with DC excitation in the rotor is de termined by the stator frequency and the number of poles. As long as effi cient, variable frequency power supplies were unavailable this meant constant speed operation at fixed frequency. There are drive applications , where con stant speed is desired or where the reactive power that can be generated with line- connected synchronous motors is an important feature; these are, besides electromechanical docks, mainly high power drives, such as for pis ton compressors in the chemical industry. Another field of application exists in pumped storage power plants, where the synchronous generators serve as constant-speed-drives for pumps in periods of low demand for electrical power, feeding water into e!evated reservoirs for later use during hours of peak demando ln this instance the type of motor is, of course, not a matter of choice but the synchronous machine is very welI suited for this duty; it is, in fact, the on!y one that could be used at a power leveI of, possibly, several hundred MW. Problems with !arge synchronous motors operating on a constant fre quency supply may be caused by their inherent oscillatory response since the torque now depends on the load ang!e, and the required start-up procedure. Asynchronous starting at fuII or reduced line voltage with the help of the damper winding and the short circuited field winding as welI as special start ing motors are common practice; more recent!y, large synchronous machines are also started with variable frequency supplied from static converters. This can be extended to variable speed drives with many interesting features. Historically, the synchronous motor fed with variable frequency from a DC link by thyratrons represented the first attempt at assigning the task of commutation to static equipment [A15, W29j. The thyratrons, gas filIed discharge tubes with heated cathodes, that were available in those days, were of course hardly suitable for converter duty. Thereafter, the converter-fed synchronous motor did not receive much attention until this was completely changed by the advent of high power solid state switching devices available today. As :;( ' (' 11 til t.!t(! s'yllopsis in Table 11.1 there are three main fields of appli ca l.ioll
rOI
II djll:lbhl,· ~peed RyJlchroJlOWi lllotor drivcs ,
330
14. Variable Frequency Synchronous Motor Drives
• Large, low speed reversing drives, such as needed for gear-Iess rolling mills or mine hoists, with stringent requirements for high dynamic performance. ln the past, these were mainly DC motors but they can now be built as AC synchronous motors supplied by cycloconverters, thereby eliminating all the design- and operating- restrictions inherent in DC machines [J1,L6, N4, R23, 84, 85, 867, 879]. • Large, high speed compressor drives up to 100 MW for pipe-lines or wind tunnels, where the induced source voltages and, as a consequence, the ca pability of the synchronous machine of supplying reactive current, permits the use of simple DC link converters with natural or load commutation. DC motors could not be built at this power rating and speed because of the commutator [G25,H5,K63,N15,Vll]. 8ince a synchronous motor does not have to be magnetised through the con verter, as is the case with induction motors, a larger airgap is no disadvantage and can even be desirable for mechanical reasons. The slip-rings for supplying the rotating field winding with direct current can be avoided by employing an AC exciter with rotating rectifiers as is common on large turbo-generators. • A third important and growing field of application are low power, high dy namic performance servo drives (usually < 10 kW) with permanent mag net excitation and transistor converter supply (MOSFET or lGBT). Their main advantage, when compared with induction motors, is the almost com plete elimination of rotor losses; on the other hand, field weakening is more difficult [B29,G35,J2,L48,L51,P28]. After the initial problems of achieving a wider field weakening range have been solved with internal magnets built into the rotor, permanently excited synchronous motors are also an option for vehicle drives . Large synchronous motors with permanent magnets are of interest for the propulsion of military ships, because of the improved cfficiency, Table 11.1; the stator phases may there be fed from separate four-quadrant converters to attain redundancy in case of faults. The analysis and design of a control system for an electric drive cal Is for a dynamic model of the motor; with a synchronous machine this may be of considerable complexity if details such as saliency and unsymmetrical rotor windings are to be taken into account; as this is not the aim of the discus sion here , substantial simplification may be achieved without much loss of validity by restricting ourselves to machines with cylindrical rotors, when : the mathematical model of a symmetrical three-phase machine described in Sect. 10.1 remains applicable. With large machines and high speed a solid st.ecl nOllsalient ro tor is the normal choice for mechanical reasons but eV(~ll a.t low spved cylindrical rotor designs with symmetrical damper windings 1\.1"<: ()ft( :ll uscd; heIH:e the lllodel derived in Sed. 10.1 n~maills IlSdll1. Ali I'xc< :pt.i()ll is lhe switdl(:d r(:lllcl a.ll Ç(~ lllo tor wit.h it;s saliellt (:OJls tru c I.i(lll 0 1' tllI' HI. ;~ I.() r H lld rot.o r; it.H Hyll chnmo ll H fllll Cl. ioll is based Oll H('«(1I( :ul.i a lly ('·(·d i,ul'. III<' tll'(li,II'I',t.(· Willdill l'.H (III I.lte nl.ll.I." .I' p"l(,fI . ' I'lt l: i 1·1t,lI n fOI' 1'IIt.irl'ly d i!'
14.1 Control of Synchronous Motors with PM Excitation
331
ferent models and is not included here; a large specialised literature exists on this subject, e.g. [DlO, L14, L19 , P19] . ln Table 10.1 it was demonstrated how the general model of AC motors can be adapted to various constraints. Larger synchronous motors usually have a symmetrical damper winding for reducing the effects of harmonics and negative sequence current components produced by the power converter; this also tends to lower the transient reactance of the machine and thus facilitates commutation. Hence, the single axis field winding of a synchronous machines may for simplicity be incorporated into the symmetrical damper winding by inserting an ideal DC voltage source UF without internal impedance into a joint three-phase rotor winding, as seen in Table 10.1 d. This simplification is adequate for designing the control, it would of course be too inaccurate for the design of the machine. Smaller motors, such as synchronous servo motors with permanent mag net excitation, Table 10.1 a,b do not normally possess damper windings but there is a slight electrical damping effect by eddy currents in the magnets; the mechanical damping of the drive can now be achieved by the controI. Since the control of permanent magnet excited machines is simpler than that of machines with field and damper windings, this case will be discussed first .
14.1 ControI or Synchronous Motors with PM Excitation Permanently excited synchronous motors are an attractive solution for servo drives in the kW-range. There are various designs, depending on the type of magnetic material used; best results are obtained with "rare earth" mag netic materiais such as 8amarium-CobaIt or Neodymium-Iron-Boron, which combine high fiux density (beyond 1 T) and very large coercive force (up to 7000 A/cm). This makes it practical to magnetise the pole pieces separately rather than have to magnetise the complete assembly in place, as is necessary for instance with Alnico-magnets. Rare-earth magnets require also much less space and , with proper design, there is no danger of accidental demagneti sation by short circuit currents; unfortunately magnets of this type are still quite expensive. Synchronous motors with rare earth permanent magnets have higher power density than comparable DC motors because the limiting effects of the mechanical commutator are absent; the power density exceeds also that af illductioll motors because there are no losses by torque-producing rotor cur n !llts lha!. nc(:d to he removed by coolinG· A comparison is seen in Fig. 14. 1, sh()willl~ til<' 011 L1il1(: of all axia.l fidd De disk motor, until a few years ago the ullilJlat.<' c ltoÍI'I ' for dyllalllic pnfoflllll.Il(·I', illld a f' ylll'hroll o llR AC servo IIlolor wil,h nt,J' 1I I'II,I'!.II IJI ; 'JI,lId (' Xl'i!.iltioll. TIl<' !I( ~ IIIO!."I' i" 1101. olll'y 11101'1' powI 'l'fld,
14. Variable Frequency Synchronous Motor Drives
:U2
lIaving lower mass and inertia, it exhibits ais o a much larger impulse torque Llt = 900 Nms as compared to 5.5 Nms for the disk motor; this, for iustance, is important for quickly accelerating an inertia load through a given speed range Llw. The reason for this large difference is the thermal capacity of t,he AC stator winding which is embedded in slots, as compared to the open conductors of the printed armature winding in a DC disk motor. Due t o their high power density and , consequently, the smaller size, permanent IlIagnet synchronous motors have in recent years evolved as the preferred c1lOice for positioning drives on machine tools and robots, where the motors ; Ij"(~ to be integrated into the mechanicalload. The problems associated with I mddash and elasticity of gears may be avoided by using linear synchronous lllOtors with permanent magnets for precision servo drives [B78, G29, H33].
1/L ,ll a x
14.1 Control of Synchronous Motors with PM Excitation
333
for external magnetic fields, they may be viewed as part of the air gap. Hence the motor effectively features a constant wide air gap resulting in a relatively small synchronous reactance which minimises armature reaction. Skewed stator
Sm-Co
Magnets
60
r-
oo L
Ej E E E
N
= l
Fig. 14.2. Cross section of 4 pole synchronous servo motor with rare-earth permanent magnets mounted on the rotor surface
o o
C')
220mm
De Disk Motor -1.!) kW :toOO min -1, 14.3 Nm I I -1 Nrn for 50 ms or I % Duty cycle
IIlill-
:\:. 1<1.;
7-1 kg cru 2
Nominaloperation Max. Torque Torque Impulse
:•.7 NlIls
'uoo
AC Synchronous Motor
1
Max. Speed Mass Inertia
6.7 kW 4000 min- 1 , 16 Nm 90 Nm for 10 s or 15 % Duty cycle 900 Nms 4500 min- 1 16 kg 45 kg cm 2
1";1'.. 14.1. Comparison of DC and AC servo motors with permanent excitation
As an example, Fig. 14.2 shows the cross section of an experimental 1.'). k W , 4 pole synchronous motor (continuous rating) with a hollow steel rotor t.o whidl Srn-Co-Magnets produced from standard size bars are attached, t.J1l1 1i rednCÍIlf:!: also the quadrature field. The whole rotor assembly is bandaged wi t.1I K('vlar fibre after the hollow parts have been filled with light-wciJ.!:ht, J1011(".()lldlldillf:!: Hlat.(~rial. Tll(·~ stator is skcwed h.v OHe slot wifith for reduci u, \ t.iI" I.orqup rippk c<\lIsed b.v t.he st.a.tor slots [L5 '1) P '2 ~I· f LII ... (·;t.rl.h P(·I'II1;Ulelill. 11l'l.f'.ndti (';\.11 II(' ("oll sid(·I'(·d as ;Ullp"l'el.llrIl R s ourc" r1 ; Ili lll(·I.II1'.\, ('()lld,ill(' r(·I;d.iv(·ly Ilil:,h ('I('dric '; JI l"C·c:i' ll.iv i t y iI.lld ullil.)' P(·l'lIh·llhilít.,v
To adapt the mathematical model ofthe symmetrical two pole three-phase machine derived in Sect. 10.1 to this constraint, the rotor circuit, Table 10.1 a, is assumed to be fed from two current sources, Fig. 14.3. According to Eq. (10.10) this produces an impressed rotor current vector
. IF · ZR(t) = IF - - eJi 2
-
I F J' 2 3 e i = - IF 2' 2
-
ç =0,
(14.1 )
which is fixed to the rotor and creates a sinusoidal ampereturns distribution that moves synchronously with the rotor; it serves as a basis of reference for the control. When further assuming that the stator currents of the servo motor are supplied by a current-controlled switched transistor converter of the type shown in Fig. 11.3, having adequate ceiling volt age and high clock frequency, the stator windings can again be considered as being fed by current sources following the commands from a set of reference values. More realistically, the response of the current controlloops may be approximated by unity gain lag elements disv. . Te - - + ZSv = ZSvRef dt
v
= 1,2,3
(14.2)
wilh L:~ isv = O, where in practice Te may be in the order of 1 ms or less. Fig. 11.-1 shows a reeorded step response of a stator current controlloop for t:hc s('rvo 1I101.0l" in Fi g. 14.2, measurcd in standstill. ln steady state condition 1 . 11(~ C\lI"I"I·III.: 1 1\.1 ')\ II.JlpI'Ol(illla.t:d.v Sillllio\llidal; (hw to tltc induced back voltage 1.11<' I'IJ : irl III'" " nill ll l'. 11,1, 1lil(llC'r f; 1"', :d .
14. Variable Frequency Synchronous Motor Drives
334 LR
iS1
iRl =/F
RR
14.1 Control of Synchronous Motors with PM Excitation
The appropriate mode of control may be deduced from a phasor diagram, describing the volt age equation Eq. (14.3) in steady state. With Wl = W = const., c = wt, the stator current and voltage vectors are
A /F
17
"2
Jwt is(t) = 3 V2 . 2 lse /F
I.
"2
o
iR3
1 ms
Fig. 14.4. Recorded step response of stator current control loop, employing a transistor converter
The stator volt age Eq. (10.50), while being of no consequence for the drive dynamics when the currents are impressed, is important for the converter designo With the angular rotational speed w =dc / dt it follows
+ L S didtS + J. w '32 L o I F
j
e
E
= Y.s
Y.s (t)
3 V2
.
Jwt = -2U -s e .
(14.5)
After simplification, Eq. (14.3) assumes the form
.1
(
)
(14.3)
t .
+j
w LsHs
+E
= Us'
(14.6)
where . 1 E=J rnWPF
-
· R S 1s
,
v
(Rs
Fig. 14.3. Equivalent rotor circuit of synchronous motor with permanent magnet excitation
335
v2
(14.7)
is the RMS-value of the source volt age induced in the stator winding by the permanent magnets, Fig. 14.6 a. The expression . () . t 3V2 3V2 3V2( . !s t e- Jw = -2-ls = -2-IseJ ó = -2- ISd +JISq)
(14.8)
The expression for the electric torque is mM(t) = ~ Lo 1m [is (iR e jE )*] = Lo IF 1m [is e-j E) = PF iS q
(14.4)
where is e-j E is the stator current vector in a coordinate frame given by the rot.or position. Since c is easy to measure, this is the preferred orientation !«'re; it. corresponds to the d - q or Park-Transformation commonly used with syllr.broIlOUS machines. 'rl)(~ mathematical model of the synchronous motor is contained in the li!';!'1. halld section of Fig. 14.5. The principIe of controlling the motor is ;:illlilar to the field orientated control of an induction motor except that now I.h(' rolor position serves as the angle of reference; hence there is no need for Õ\ J1l1X model. The drive should of course be operated in such a way as to lIIake best use of the motor and converter. Converter
Control with microcompuler
-------f--
-t--
with currenl control
PM exciled synchronous motor
Lag
~F
comoenS8tion
/'11\)(/011
mL
n'"/I.,/lI/{
represents the stator current in rotor coordinates, with ó being the current load angle; ó = O is the no-Ioad position. The phasor diagram is drawn in solid lines in Fig. 14.6 b for ó = ~ which is the optimal mode of operation, where the motor produces maximum torque for a given stator current. With a synchronous motor operating on the constant frequency supply, it would correspond to the static stability limit, but this is of no consequence here as the motor is self-controlled, similar to a DC machine where the commutator is fixed to the rotor. Clearly, ó = ±~, ISd = O is the best choice as long as the converter can supply the necessary voltage, i.e. below base speed. Direct field weakening is not feasible with a permanently excited machine but a similar effect can be achieved by advancing the current vector beyond ó = ~, i.e. introducing a negative direct current component. This is indicated in Fig. 14.6 b in dashed lines resulting in a decrease of the terminal voltage USI. As a consequence the maximum value of the quadrature current ISq may have to be reduced in order not to exceed the limit for the total current,
1"_,,, [1'_~~.~I.(If - ~U" \ r' - ~ fi
i ..
Is
1+}(I)To
i' dll," - O
11,·/, 1" ," I,/h lfl''' OJ
+- - - -
,r lt'>. 1·1. (,. (:",,1.,·,,1 "f "'y "l'ill'''''''''"
~ 1 '1I11.l1 , I ~ '111/11/,1,;1"/
+
""il... r wJl.IJ P(·""III.',, '"I. III /lt( ''''!'
~
11" t••/ ,-,N)/(llIml" 8
oxc·il.stl.io"
= /I~d + I~q < Is max ,
(14.9)
which leads to a corresponding reduction of torque, a characteristic feature of opel'atioll at reduccd field. Tlw opl.illla) coutrol procedure is thus as follows: Operation with ISd = O, I/ .~·"I · I", ,,,,,,,, 111' 1.11 I.I\(~ v()lt.a[~(~ lilllit., which is rci'tched ato the base speeo; a rllrl.l\C ,)·l lIc ' ,n,,, \,·l ll li l'c·,·d ":111 1.'111'11 1)(' ;I,.hj,·v(·" II,Y Hllirtilll': 1.111' (',111')'( ' 111, illt.o 1.111'
14. Variable Frequency Synchronous Motor Drives
336
RS 'Sd r-
Zs
(~ ~:
/".
RS
júlLs1Sd l
)
/
I / 1 I / /2 s 1s1
fi
!Ls
19 = arctan
I \ /-S
Ls
W'
I~ull. -....1~ - SI
"'-....
ôl
ct>F
f max
wLs 19 1 = arctan Rs .
!Ls 1S=ISd
ct>F
.!.S=jISq 'Sq< O d
c
Fig. 14.6. Phasor diagrams of PM synchronous motor, (a) Equivalent circuit diagram, (h) Phasor diagram showing effect of field weakening, (c) Phasor diagram during regeneration, (d) Phasor diagram at maximum no-Ioad speed
unstable region, 181 > ~, while maintaining constant terminal voltage. ln view of the large effective air gap with rare-earth permanent magnets, this method of "field weakening by armature reaction" calls for a large stator current in the direct axis and, hence, is a much less efficient procedure than would be possible on a synchronous machine with a De field winding; however, as a short time measure it may be acceptable. As seen in Fig. 14.6 c the need for field weakening is likely to occur mainly in the motoring quadrants but, ia principIe, it is applicable to braking operation as well. A possible solution to the control problem is sketched in the lower part of Fig. 14.10, where field weakening is initiated by an auxiliary control loop with the aim of limitiu[f, the magnitude of the stator voltage. The degree of field weakening that may be achieved by armature reactioJl can be estimated from the phasor diagram, Fig. 14.6 b. By initially assullliJlj{ a purely quadrature current, the voltage components are
Us eos 19 ( / ' : : i i II ,ii
= fi) + Rs rs·,! , t ,) /, ,": 1,'{I/ ,
(14.11)
where
j úlmax Ls 'Sd
ct>F
(14.10)
US 1 = Us cos(19 1 -19)
RS/Sd
li =-~ 2
wLsIs q E+RsIsq
By now adding a direct component, ISd < O, while maintaining ISq and hence torque, the volt age phasor U S moves along the dashed straight line in the indicated direction. The minimum value of the volt age is reached when the phasor U S is orthogonal to this line, hence
'Sd
b
a
337
where the voltage angle 19 depends on speed and load,
IZs ISd/ U
.!.S
14.1 Control of Synchronous Motors with PM Excitation
(14.12)
There is no point in advancing the volt age angle beyond 19 1 , as the stator volt age then begins to rise again. The additional direct component ISd, which appears as a reactive currenl. component at the motor terminaIs, causes an increase of the stator curn~lIl. Is; the current IS 1 for the condition of minimum stator volt age is founel f!"OllI the related triangles
I S1 =
E2 + U§l - 2E US 1 cos19 1 I E2 + U'§ - 2 E U S cos 19 Sq .
(14.1:\)
Since the stator current is limited by the converter, this calls for a corn~'· sponding reduction of the maximum quadrature current. The following parameters, taken from the 1.2 kW servo motor (Fig. J.1 ,:~), may serve as an example of what can be expected with this method [I ,r, II, At the base speed (2000 l/min) and at nominal torque (5.7 Nm), the J"(~bl.iv(' voltage across the stator impedance is w Ls ISq = 0.139
Rs ISq
E
= 0.066,
which results in U s/ E = 1.075 and 19 = 7.5 0 • By adding a negative Isd-component and adjusting it for minilllUlll :; La/.. ,r voltage , the voltage angle becomes w Ls 9 =- arctií,n - Rs
'I J
= 64.5
n.nd t1w sl.a.l.nr voll.agl' is (1 ,' ,' 1
II "
/ j'
l /I'
"
"1 111 1'0, \
o
,
r edll(",( ~ d
.,'1)
1.0
(J .r,.~ 1 1 "
14.1 Control of Synchronous Motors with PM Excitation
14. Variable Frequency Synchronous Motor Drives
:138
i\.t the sarne time, however, the stator current, Eq. (14.13), ri ses to
IS l ~ 6.0. ISq The torque and speed in this assumed operating condition remain unchanged with the effective stator impedance reduced by a factor (Ust! Isd/(Us/ ISq) = 0.091; this permits access to suitably defined operating points in the low \.orque/high speed region, as long as the volt age and current limits of the converter are not exceeded. ln view of the high copper losses this procedure is of course restricted to short transients at light load which is common prac "ice with DC or induction motor AC drives. As a limiting case the situation depicted in Fig. 14.6 d may be considered, whcre the stator ampereturns are opposing the rotor excitation. This would result in the maximum speed obtainable by controlled armature reaction at zero torque, Is = Is d < O.
From the volt age equation
u~ = (Emax
current is 8000 l/min. Better results in the field weakenig range are achieved when choosing different stator impedances in direct and quadrature axes. Fig. 14.5 contains a block diagram of the rotor orientated control scheme in rectangular components but the control can of course aIs o be presented in polar formo This is shown in Fig. 14.7, where a microcomputer performs alI the control functions up to position control; operation at reduced field is achieved with the help of a suitably defined function 8Ref = f(w).
2000f:m;~~ ___
ú)--
-2000
w W i I. I!
~-tRsIsd)2
E
Ifo!71S:;
I
(14.15)
+ (wLs ISd)
tlw numbers listed above this results in
t• ffi
' 's .
28 J -;ii .......__
= 1.25.
higher speeds can be obtained by allowing larger currents. With pnrlicular servo motor described, the no load speed at five times nominal
(lr mllrsc ,
!.II<:
~ Converter and
OlA
15
motor
'1711+.____
• t
.
'S1
100ms
1. .1
Fig. 14.8. Speed transients of microcomputer-controlled synchronous motor at no-Ioad with six times nominal current (a) speed reversal, (b) acceleration intofield weakening region
The transformations from stator to rotor coordinates and vice versa now call for a simple modulo 27r addition of é. All the inner control functions are executed in 0.6 ms-intervals which is adequate for stator frequencies of up to about 200 Hz but a further extension of the frequency range could be achieved with the help of signal processors. Transients that have been recorded wit.h the 1.2 kW permanently excited synchronous motor at six t.illll:H /loJ!lillilI cnrl'(·nt. àre displayed in Figs. 14.8 and 14.9. The control is p(:rforllwd I,)' H III k ro("ol!l pn t.m' wit:h a (:OU I. rol ~trlldllrc similar to the O)W
con'rnllar
F lu . 1'1.7 . (:t"ll" ',,l'lf
Rei
-28
b MicrocompulSf
(
.
I
(;.)"",,/w
•
isqRef
800: I-m;~(
rol\ows with Emax/ E = wmax/w the maximum no-Ioad speed, referred to base specd, at nominal current W rn ax
t
W'
a
(14.14)
+ W max Ls ISd)2 + (Rs ISd)2 ,
339
1I'y llt ' III 'I HHII H I IIlCd,llr rl ( 11' V 11
dl'lvl'
iII
pnl ht {·()()r~líll ll. l,t·!I
:;]IOWII iII
Fi",
1·17; I,,'wc'v('r ,
:qwt"illl'
/I.ct...pl,ivc ·
li "rt.Wil.l1'
('(·at.u t c·:: hl\.Vc ·
],C'I"II
14.1 Control of Synchronous Motors with PM Excitation
14. Variable Frequency Synchronous Motor Drives
:110
341
I
n min -1 5000
Step size US3
8 I 16
32Rev
I I I PM synchronous motor I .-e____~ I I I I ~ A USb
ú)
--t
~ 500 ms-----+l
t
e
Rev
·jE
32 USa
I I I I
16
8
4
r------
I
USdRef
Fig. 14.9. Step response of time-optimal position control with synchronous motor servo drive at no-Ioad
ejE ú)
aclded to the speed- or position-control function to combine fast response with aperiodic damping [L51]. The speed response of the drive is limited, besides the inertia, by the llIéLximum current which the converter can supply. Given sufficient stator c1lfrent, the motors have very high short-time overload capacity which is Ílllportant for servo drives designed for intermittent duty, Fig. 14.l. Tile control circuits in Figs. 14.5 and 14.7 contain current controlloops in ::I.al.or coordinates, producing approximately impressed AC stator currents. 'I '11<, rderence signals are sinusoidal in steady state which calls for careful 1.1111 i II!'; of the current controllers with regard to undesirable phase shifts. An ;dl.cruative and usually preferable arrangement is shown in its basic form in I"il\. 14.10, where the currents are controlled in moving rotor coordinates; this \tas the advantage that the controllers for is d, iS q are processing DC signals ill steady state. The effects ofthe induced voltages may again be compensated I)y L"eed forward at the outputs of the current controllers; the control can also Iw cxtended for field weakening. As the phasor diagram in Fig. 14.6 b and the parameters of the experimen.. Lal lI1otor indicated, the increase of the stator currents in the field weakening r!'gion is due to the small stator inductance Ls, caused by the relatively largc dh:ctive airgap as a consequence of the surface-mounted magnets. This enu Iw illlproved with different motor designs, where the rotor is anisotropic, sim ilar Lo a rotor with salient poles, thus increasing the longitudinal illductallGC I'S,1 colllpared with tbe qlladrature inductallce LSI' [B26,ll28,D29,J2]; p osKi illl: ÍUlpkJlWIlLatiollS of this idca are seen in Pi g. 1-1.11 a,b. 'I'lte lll al~ U(;Ls lJI:l,y ..il.ll('l' 1)(' NILI I'acn IIIOllllLI:d wiLh UI(' :wiril d,rop.v cltllscd h.v UI(: i 111.1 'I'Jlal :·dt:l./)(" 0 11" 1.1)(' I'(d,I)I' I» I.Iwy 1"lltld 11)1' !Ir);I.III':I,d illl.clllill!y lf\,u.., II ~w ) lo·m ) I\iI:1HI ; iII l"d,11
USqR.f
Voltage source converter
Ú)Ref
Speed control/er
Control
Fig. 14.10. PM-Synchronous motor drive with voltage source converter and current control in rotor coordinates
cases laminated rotor constructions would be used. These designs permit an easier displacement of the impressed flux by the direct stator current com ponent is d into geometricalleakage paths. Of course, a detailed modelling of the machine is much more complex but in priniple the result would cor respond to the simplified model in Fig. 14.10 where different values of Ls d , LSq in direct and quadrature axes are assumed, caused by varying cross sec tions of the magnetic paths and, possibly, saturation. ln the design shown in Fig. 14.11 b1 a cage winding is included for self-starting of the motor on a constant frequency supply. With these measures PM synchronous motors may exibit a substantial degree of field weakening without excessive stator eurrents; this is necessary for instance on vehicle drives, reducing the need for a mechanical gear change. Besides the PM synchronous motors having sinusoidal stator voltages and eurrents and smooth torque, i.e. low torque ripple in steady state, there are also ll\(»)'(: ccoIlomica.l designs in the form of "brushless DC motors" , exhibit illg t.rapczoid;t,\ Illlx dCllsity distribution and phase voltages [839, W14]. The :,t.al.(lr C·IIITI'II!.f; !t;\VI' ju prillciple a sl[u;ne waveform, as in Fig. 8.19 d, where 011<' ,,1 11111" ,' (lrI" 'III, \:;
'!,('rII
at, 1\.1I'y I.illll' ;
l'iI.lIHI'c!
!ly t.!tc' finite COlluIllltat.ion in
14.2 Synchronous Motor with Field- and Damper-Windings
14. Variable Frequency Synchronous Motor Drives
:)42
tervals the currents are of course continuous, i.e. the corners are rounded ofr. The main advantage is that there is no need for a pulse-width-modulated converter and maximum power density is achievedj however, due to imper fcct commutation there is considerable torque ripple making these motors less suited for high precision servo drives.
343
harmonics and negative sequence ampereturn waves, not for starting the motor because this is now achieved with the variable frequency supply. The basic circuit of this type of drive is shown in Fig. 14.12.
S/a/or
6 - pole motor with
magnets on rotor surface
a)
References Magne/s S/a/or
-axis
Cage
winding
, ~.
, aifga.p~ , . . . .;,.
.
b1)
M.agne/s in
)
b2)
b 3)
x' .• ·l ...· : ' ·
-
,..
Fig. 14.12. Variable frequency synchronous motor drive with cycloconverter 'i .
....
4 - pole motors with rotor- internal magnets "'i~. 14.11. Cross sections of PM synchronous motors with :\Ilisol.ropic rotor designs for extended field weakening range
L1.2 Cycloconverter-fed Synchronous Motors with
Field- and Damper-Windings
/\ \l( )Lher type of adjustable speed synchronous motor drives employs machines 01" couventional design having in the rotor a symmetrical damper winding aud a. single axis field winding fed with direct voltage through slíp rings 01: rrolJl :Ln AC exciter with rotating rectifiers. At low speed, for instance bdow tOO I /llIill, Lhe stator can be supplied by a cycloconverter employillg Iim~ C()1ll11l1l1.:üioll which makes this scheme particu!arly aUractive for high pOW(>f :l ppli(";d,i()ll~;, as !Ollg as thc [requency mllg(~ of t.hc cycloconverter slIllire:j OH' In r \',"(' 11I11IIhcr oF t.hyriK!.or brandlcs i~ !lOt. ()IJj u<:t.i(lnalJ! ~: a!. hi gh p()w( ~r whnn '
1'1f.1·n ll .. ll. h v ri :iI ."r:: IJlil':ht. ()i.lJl"rwi ~;( ·11(' 111'1,,,, ·<1 . '1'111' nll,OI" IIIIH Il lI ill/ (I,' lI.X iH lidei w l od l lij/. [III" II, ::vl'II'II"I.JÍ ,·,d 01 0 1111"'1 Willdillj ( w hk ll 1111 1 ,,( 'ti 1'' '1 1IIlI'pro 'llnilll\
Typical applications are large gearless drives, in the past dominated by DC machines supplied by dual converters (Fig. 9.4), for example for reversing rolling mil/s, mine hoists, ore mills and rotary cement kilnsj at lower power, elevators or direct wheel drives on excavators in open-cast mines are also possible applications. The drive permits 4-quadrant operation with smooth transitions including standstillj hence it is perfectly suited for speed- as well as position- control requiring rapid response. The torque harmonics due to the cycloconverter are small and of relatively high frequency. An additional advantage is that cycloconverters with air- or liquid- cooling are proven de signs from the DC era and are readily available up to MW rating. A detailed analysis of the drive calls for an extended machine model in cluding the converter that could not be dealt with manually or by intuition but would require a computer simulation [M17]. For qualitative purposes, however, the model derived in Sect. 10.1 still remains useful, as seen in Table 10.1 d. Rence we assume that the motor with a nonsalient cylindrical rotor has a symmetrical three-phase rotor winding into which, according to Fig. 14.13, an ideal source of direct volt age UF is inserted as field power supply. Since this controllable voltage source contains no internal impedance, the damping efrect of the rotor winding remains unimpeded and the symmetri cal rotor winoin g is used for DC excitation [L42]. The simplified model is not ofFcrillg the sam e freedom for choosing parameters as a separate winding arrallg<~IIj(~ II[. hut. lIlay be adcquate for the design of the control system. To a.da)'!. 1.111' Illa!.lu 'lllaI:Ícallllo,ld ,!priv<'d in Sect . 10.1 to this constraint, w,' il l" ".ln d, ·lill" ",11.h ·'Y 211;:\ a vl'd.OI (Ir ,' x1.1T1I ali y appJi(~d I'ot.or vo!t.a.f'/'~;
14.2 Synchronous Motor with Field- and Damper-Windings
14. Variable Frequency Synchronous Motor Drives
:1'14
.
.2
+ UR2 eJ'Y + UR3 e)
Y.R(t) = URl
~ (UR2 + UR3) + j
= URl -
'Y
V;
(14.16)
345
equation and the mechanical equations for modelling the drive dynamics, Eqs,(10.51-1O.53),
(um - UR3) . Up
di R · = R R '!cR + L R dt + L O dtd
~
=
(' - j e) '!cs e ,
(14.20)
I kcause of the short circuit between phases 2 and 3, UR2
J
= UR3,
Itolds, hence
~ Lo 1m [is (iR ej é) *j -
(14.21)
mL ,
dE
Y.R(t) = URl - UR2 = up(t) .
'1'llUs the rotor voltage vector is real, being identical with the applied field vo ltage, Fig. 14.13.
URI
=iF
iRI
Induced voltage
ºSI
LR
RR
Rotor flux axis Rotor axis
RF
UR2= uR3
jmR
~ uF
íR2
Stator axis
3
='2 RR
RR I"if~' 14.13.
dt =
(14.17)
Symmetrical rotor winding Fig. 14.14. from a voltage source system
Definition of coordinate
(14.22)
W.
The design of a control scheme in field coordinates calls for the selection of a suitable frame of reference. ln the case of the synchronous motor with permanent magnets, Sect. 14.1, the rotor position € was the natural choice as this was the orientation of the magnets producing the rotor ampereturnsi however, the situation is different here, where the induced currents can shift the ampereturns-wave with respect to the rotor, Fig. 14.8. It is therefore appropriate to employ a fiux vector as with the induction motor. A suitable choice is the rotor fiux, defined in Eq. (12.22), i mR
= is + (1 + IrR) iR ej e = imR ej e .
(14.23)
This makes sense because the rotor winding in Fig. 14.13 has alI the features of an induction machinei the only difference is that in steady state the mag netising current vector imR is in synchronism with the rotor, (! - € = const., whereas it was moving with slip frequency across the rotor of the induction motor, Fig. 12.11. Eliminating iR from Eqs. (14.20,14.23)
1)( : ( ~ x:citation
\ViLl,
iR
TIl!' rotor currents have mean values, to which transient terms :; lIpl ~l illlposed. Considering the isolated neutral of the winding, we find
"Íu (t)
Ip + i ) eJ'Y = (Ip + i Rl ) + ( - 2 2 + i R3 R2 . + (-I p
./) + J. 2 J3 ('/ = 3'2 (Ip + 2Rl 2m
-
./)
) '2 e)
dimR TR ~
are
dimR. TR ~ + ZmR
'Y
d
dt ((! -
'l'lte JIlcall of the rotor current vector is also real,
. 'lUO
3
3 Up
= '2 11" - '2
~RR
(14. H»)
1'11l. IlJld (~ r dYIlamic conditions the vector can be deflccted by indl1ced currelll. 1',1'III pl)Jl( ~ IIL s. I k s l. ['1'"uILs '.1'\'
1"'11'1.
or
t.IlC
drive c()ul.rnl fIIay
h( ~ (' X:I)( ~d,ed
ir tll(~
sLaLo1' CllJ'lTlILH
H' I ~ f~ill , ' IIIT(' 1I1. i lUIII'CC'( 1 Ily \.II<' ('yrl()((IIIV~·I· t' · I· wil. h t.lw 1tC'lp III' f f\)i l, CII!' n .,oI.r.. 11III' l[i ll , lI ll wa.li 11 [" )Wll iu :-\",'1,. 1:1.1 . '1'1 ';11 1 \' I ~W I I UIC' l'ul.1I 1' vl,H,/[I\"
Up j e + (1 + Ir R ) RR e
)€
. ( Up = ZSd + 1 + IrR) RR
= WmR -
(14.24)
,
(
(14.25)
cos (! - )€ ,
W
=.
1T
= is ei (Ç-(J) =
iSd
ZmR
Up
RR'
. = '!cS
and splitting in two real differential equations yields
(14.18)
ZR3 .
+ (1 - J. W TR) ''!cmR
R
[i S q - (1
+ IrR)
UP R sin((! R
é)]
(14.26)
where is e-·j
(!
+j
i Sq
(14.27)
is :~g aill ti II' sl.:l.I.or Cllrrent vector in rot.Qr {lIlX c()ürclinates. Sillc '(' 1.11<' wc\...,· i,,"1I \IIoves ill H\.c~:i.dy ,,1.aU' sy lldll'lHlOIISl y wit:h til<' rolor , \.11<' I·il,.lel. le ll lld l:id o o" I';". (l /I. 'l (i) VIl, lI i N llI' ~ ,
346
14. Variable Frequency Synchronous Motor Drives
isql Hoo = (1
+ UR) ~:
14.2 Synchronous Motor with Field- and Damper-Windings
(14.28)
sin(º - ê)IHoo .
The equation (12.27) for the driving torque remains unchanged, mM
= ~ Lo 1m [is (iR ej ê)*] = ~ (1 -
u) Ls imR iS q
I Converters I
Field I
supply I
u~
~
II
I
•
347
Synchronous motor
(
= k imR iS q • (14.29)
The equations describing the synchronous machine in field coordinates are graphically represented by the block diagram in the upper part of Fig. 14.15, containing once more the transformations from the stator- or rotor- orien tated coordinate system into field coordinates. The diagram corresponds to that of the induction motor with the exception of the field voltage UF which, alter transformation into flux-orientated d - q-components, is added as an additional input signal at the appropriate points. ln practice UF may be sup ]llied by a separate two-quadrant converter with closed loop volt age control, lhus underlining the characteristics of a low impedance source. lt would of course also be possible to introduce a control loop for the field current to eliminate the effect of changing field winding resistance due to temperature variations. The drive would then respond faster in the field weakening range, but the damping effect by transient currents induced in the low impedance held circuit might be impeded; also the model of the machine would have to I JC Illodified. ln contrast to the PM excited machine in Fig. 14.6 there is no need to apply stator current in the d-axis for field weakening, since this can now be arhipved more efficiently by reducing the field voltage rather than increas illf~ I.hc stator current. The need to reduce the field becomes manifest again wll(~1l lhe magnitude of the stator volt age approaches its limit as sensed by a I i Jlli I.illg voltage controllerj alternatively field weakening could be performed wil.h an imwcontrolloop, the reference of which is generated as a nonlinear J'lIllcl.ion of speed, imRRef(W), Fig. 12.14. 'fIte flux angle {! required for modulation and for computation of i m R, is " alld mM is again derived from a ftux model, processed on-line by a mi crocomputer. As in case of the induction machine, the model may be defined ill ~ I.ator or in field coordinates, Fig. 12.16 a,b. Computational advantages are gained by taking the second approach, referring to Eqs. (14.25, 14.26), whidl descri be the magnitude of the flux vector and its position relative to 1.Il<' rotor. This is represented by the block diagram in Fig. 14.16 which is a rop'y ()r l/te motor model in Fig. 14.15. Clearly, with UF = O, the ftux mode! ()f 1.1t<' imludion machine, Fig. 12.16 b, results. The angle º- ê is approximately (,()ll~;talll ill st.eady state. 'I '/te ;u:r.uracy with which the stator cllHcnls are tmdcinp; thc refcrenq:1i ;\'I1;ain h t: c()lIsi(I(~rn.hly illlproved uy (~ mpl()yil1f~ voll.ag<: fCüd- fofward; il,fI i:i t" ('auc<'! thl' V()!1."~I_~t' (' .... h( 'hilld 1.l1(' 'ntll ~ ienl. n 'li.t'.I,llI 1('(: whidl li 1Ilcllll'('c! I, y 1.1,.. 1I1 : If ~ nd. lní lll', (·ul' r l'II!., Lo. UI<' 11IiI~1I UI ! X . IIJlty
pllrpOIi('
iS! Rol
I
iS2RoI
I
iS3R611
Fig. 14.15. Control of synchronous motor in field-coordinates
The stator voltage was derived in Eq. (12.45)
.
1is (t) = Rs 1s
+ u Ls
~s
dt
+ (1 -
u) Ls
~mR &'
(14.30)
which in steady state, Le. with WmR
= Wl = W =
const,
imR
=
const,
is
=
const
assumes the form llS(t)
= (Rs + j W u Ls}is + ~s(t) ;
(14.31)
hence the induced voltage is §.s(t) = j W (1 - u) Ls i mR .
(14.32)
If the control functions are performed by a microcomputer, the sinusoidal
three-phase components of ~s can be easily computed and fed into the current loops behind the current controllers as is shown in Fig. 14.17. Experimental rcsults were seen in Fig. 13.8. If the rurrClIt controllers are of the analogue version, being part of the i>ow(~r (·(llIV(:rl.(:1's, the schenw in Pir~ . ] 4 . .1. 7 a wOllld call for a second DIA ('OIJVmi,l lI ' Um!. ('1 \.11 IJ<: OJllit.t: ( ~ d hy IIlixilll': 1,[1(' t-/igua!s wil.hi.ll. t!t(: IIlinoco)ll PIlI. ... I', 11·1. :1\" 'WII i II '<'ii', . 14.17 h. ()J' ('(Jllf li(', jll'l'l'l :<'I. 1':I.II<:dl :t.l.ioll (lI' I ': : iI! Il(ll.
14. Variable Frequency Synchronous Motor Drives
:\t\tl
14.3 Synchronous Motor with Load-commutated Inverter
349
An altemative approach to the controI shown in Fig. 14.15 wouId again be to controI the stator currents in fieId coordinates, according to Fig. 12.27 where the current controllers are processing De signaIs in steady state.
t'::J
p-e
i. ;,-
e Fi~.
14.16. Dynamic model for computing the rotor fiux of synchronous machine
I
O IA
<"SI
Feed - forward
Current
Fc
.
-J-: I Converter and motor
Mlcrocomputer .I
Fc=G cTs+1 Ts I
ti: .. I
Current /". , :;(1/1
Lêw dlLag
II
F ig. I 'J .] 7. Current control with voltage feed-forward
with any of the schemes since the nonlinear control dynamics of the the point of compensation from the actual entry point of !.II(' di.~l.llrhnllcc. 3till, considerabIe improvements are possiblc, as was ShOWll j II I,·jl~.. 1:\.8. Thr residual lag of the curnmt loops lIlay be furt-her reduced 1,.\, i 1I~:( · I'I.ill!'; [('ad . laf.!; liIt.c:rs :1t the refcrcllcc inpllts of the currcnt controllcrsj I.II il-! íH i lldk:iI,('d ilJ Fil~ . 11.17, pos,: ihlc
(·o llv(:rl.( :r s(~i>arate
14.3 Synchronous Motor with Load-commutated Inverter (LeI-Drive) The variabIe frequency synchronous motor drive with cycloconverter sup pIy discussed in the preceding paragraph is suitabIe for Iarge drives with high dynamic performance. 8ince the cycloconverter is line-commutated, onIy converter-grade thyristors are needed, although in a Iarge quantity. On the other hand there is the frequency limitation of cycloconverters; with a 50 Hz supply and 6-puIse circuits the range extends to about h = 20 H z. This is sufficient for a rolling mill motor having a base speed of 60 l/min and a maximum speed of 120 l/min; it could be realised with up to 20 poles at practically any power rating [84,879]. However, when Iarge high speed drives are needed, for exampIe 4500 l/min on a compressor for a pipe-line or a bIast fumace, a maximum stator fre quency of 75 Hz wouId be needed even with a two poIe motor, so a cyclo converter can offer no solution; neither could a De drive have been built for this duty. The rotor of the synchronous motor is made of solid steeI, similar to that of a turbogenerator. The simplest way to eliminate the restrictions imposed by the line fre quency is a two-stage conversion, inserting a De link for decoupling of the line- and the machine-side converters. Fig. 11.14 showed such a circuit con taining only 12 thyristors; the machine-side converter can be further simpli fied with a synchronous motor because the stator winding contains a volt age source fs in series with a transient impedance, Eq. (14.31). Hence the machine-side converter can be commutated by the Ioad without resorting to forced commutation as was necessary with the induction motor; in other words the synchronous machine is operated in the overexcited region so that it can supply the reactive current neeeded by the phase-controlled machine-side converter. The basic circuit of a Ioad-commutated inverter drive (LeI-drive) is seen in Fig. 14.18. This leaves the problem of commutation at start-up, where the induced volt age is insufficient, but this can be circumvented by intermittently bIank ing the link current with the heIp of the line-side converter. This produces Lorqnc pulsations, but there are many applications where smooth operation ill tlw very low speed range, say beIow 10% of nominal speed, is not neces sary; examples are the turbo-compressors mentioned before, also gas-turbine 01' PlllllP(~d storage sets, which are started with the synchronous generator a.diJlI'. a.s
ii. 11101.01'.
()III.Ni d('
a.
low s p ~'t:d l'l1,1I J«: , wli!'!'(· UI<' (' IIITl'lIl 'i I> IIIllsl. ),(' IIlad(' »11"\\1 (·"IIIllIlll.lll. l. 'l1 01' 1.11" "' /ir llÍl ... lIi d,· (·""V(',·I.<-I', I.! ... d" lv('
V('ry
illl.('l'I l liu(·,d, 1,,,
:350
14. Variable Frequency Synchronous Motor Drives Supply Shunt Thyristor
Line-side
Machine-side converter
14.3 Synchronous Motor with Load-commutated Inverter
351
synchronised to the variable frequency source voltage ~s which is perform ing the commutation. Outside the commutation intervals the stator current, flowing through two phases of the motor, is current-sourced by the DC linkj its magnitude is maintained by closed loop control via the line-side converter. The stator current has approximately the sarne waveform consisting of 120 0 blocks as seen in Fig. 11.15 aj one stator phase carries no current. The distorted waveform of the currents, which also produces torque pulsations, and the need for low impedance commutating paths make it desirable to provide continuous damper windings which assist the formation of a smoothly moving rotational flux wave, as was shown in Fig. 12.30 and 12.37. The combination of • current sources in the stator (except during commutation, where two ter minaIs are short-cicuited through the converter), and • low impedance rotor circuit
References
F ig. 14.18. Six-pulse converter with De link ror supplying a synchronous machine (ú
,
'n
~
uo > O "
I
max
• __ ~_
Uo> O , \
max
Pulsafing forque
'
(ÚS
"
iomax' uo
~ -' \
m
, .,. Uo < O
,,
,;--'Omax
• permit safe commutation, • result in minimum link current for a given torquej the line-side converter would otherwise operate at an unnecessarily low power factor.
.)
Fig. 14.19. Operating range of synchronous motor drive with direct current link
can operate smoothly in all four quadrants of the torquejspeed plane, as is illust.rated in Fig. 14.19. During regeneration the voltage tiD is reversed by ddayed firing of the line-side converter whereas the direction of rotation is dctcnuined by the sequence of the firing pulses for the machine-side converter. 'l'lw voltage UD rises with the speed up to the li.mit given by the converter, II furt.hnr incrcase of speed is then achieved by reducing the field currcnt, i.e. I.y lidd weakeuing as with a De drive. The line interactiow; of this tyP(~ (II' drive ar<: ó\.I.~() slt\lilar to th()s(~ of a convert.cr-fcd reversihle DC drivn, Chapo ~. 'I'It" I ,II :-l i( ' 1110,11' of operat.ioll 01' (.\11: 11 tncli 'i 1I(·· jo;Jd,· COI\V(:ft.,·L' COI'I'URI)(llld:, Ir,I)
indicates that a detailed analysis of the system is complex and could only be handled by digital simulation [M17]j however, by approximation the sarne mathematical model of the machine can be employed as in the preceding section, Eqs. (14.20-14.22). The flux vector, Eq. (14.23), the induced voltage, Eq. (14.32) and the computational model for the rotor flux, Fig. 14.16, also remain valido Despite this similarity of the plant structure the field orientated control strategies illustrated in Figs. 12.33 and 14.15 are not directly applicablej this is so because the stator currents cannot be determined at will, as was possible with the current source converter employing forced commutation or the cycloconverter with line commutation and current controI. Instead, firing of the thyristors in the machine-side converter must now be so timed as to
tI, !! 1 d" Il('l ilwd iII S,'~'f.. ~ . : I, ,'x'l''''pl, 1.1., 11,1', til,' 1Ir'Í1I 1', 01' I.Ii ,vrli1torll 11111:11. I" ,
These are overriding conditions which can best be fulfilled by always operat ing the machine-side converter at the limits of the control range, i.e. • zero firing delay (a = O) when regenerating or • maximum firing delay (ama",) in the motoring region. The firing delay a is defined with respect to the induced voltages es behind the transient impedance. The two modes of operation are sketched in Fig. 14.20 a,b. Clearly the commutation interval and the fact that the commutation must be completed before T = 7r, prevent operation with purely quadrature current. The situa tiol\ is crit.ical in motor operation with the machine-side converter serving as ali illvI:rt.er h ec;w ~p the extinction angle" explained in Fig. 8.17, should be "'II;dl j(\ '"'d"l' 1101. 1.0 J'(:dllce tlw t.lll'qllC IIlHwc(!Hsa,rily Illlt OH th0. other ha.no ii, 111111 11. h" 1111',,;,·,
"IIIIIII~h 1.11 II,I,I
n,, 'IIv,'J'Y itf I,I\( : 1l1lt.1 ((liJlI ~ f.hyl'iHt.itl'. Ir
352
14.3 Synchronous Motor with Load-commutated Inverter
14. Variable Frequency Synchronous Motor Drives
a=O
o r
I
~
'"<
I / ! : '.
'. ""'
UO
•
't
a
I
~" ~ a
S1
/
i\
~
wt
='t
max
Uo>O Motoring 't
b Fig . 14.20. Current and voltage waveforms of machine-side converter with natural commutation. (a) regenerating, (b) motoring i.Iw extinction angle becomes too small, a commutation failure may be the colIsequence; while this would not constitute a major disturbance, with the lillk reactor preventing excessive rise of the current, it would still cause a :;cvcrc torque transient until the current control loop has restored the link l"1ll"rellt and the inverter returns to its regular firing sequence; the incident is Sillli la.r to the misfiring of a six-cylinder combustion engine. Another problem is the synchronisation of the machine side converter; t.illlÍug of the firing circuit with respect to the terminal volt ages of the machine would be unsatisfactory since these volt ages are highly distorted and out of pll
fs (t) = (1 - 0") Ls
ili mR dt = 1!s(t) -
.
Rs "'-s -
O"
Ls
~s
dt '
(14.3:1)
wltich cOllld l,e evalllated with analogue eircuitry or a microcomplltcr. 1\ opt.ioll is t.o use the flux lllodd iJl Vigo 111. Ili, wlw)"(~ l.wo orl.hogon a l :ll. nt.or (. I I rrl '11 Ls 'i s " , 'i s /" t.h(~ '.t lll',ular I'O(,Ul" pOl:lit.ioa f' iwdt.hc lidd vol(, Il.I ~t> "/1./,> ['( >111"' 1" '111, 1.111' illplll. i1i l':naINj t.h i H wOllldl d\'lir d Ld,Y f"f>qllin> ,t Itli('l >()('or ~ ll)lII > , >q 1,,1 " ~ "(,III i f!f', I,II!' \llI l i"lI :i lIuld i l\('rtr>rlll,l(' l.If 'll lI 'I'ft", [iI II I. "r UI<' l.wtI Ilfllll' tl Hdl\ "l ~ (~cnlld
353
was chosen for a laboratory drive, as it seems to be more straight forward and less sensitive to uncertain parameters of the motor. If the reconstruction of the induced motor voltages es is of sufficient accu racy, a c10sed loop control for the extinction angle could aIs o be envisaged, as is common practice on high voltage DC transmissions (HVDC). However, the draw-back even of sophisticated nonlinear extinction angle control schemes is that they cannot entirely exclude commutation failures because there may always be late disturbances that are not taken into account when the firing instant is determined [F21,77]. The problem with every extinction margin control is that only hindsight can tell with certainty, whether the previously calculated firing angle was adequate. A major difference between the firing control schemes for the line- and the machine-side converters is that the latter has to function with both directions of sequence and over a wide frequency range; this is a strong incentive to employ digital methods, where a variable clock frequency may be used to chang e the slope of saw-tooth signals etc. The commutation itself is nearly frequency independent, as long as the resistance of the commutation circuit can be neglected, because the driv ing volt ages increase with speed and the time available for commutation is reduced inversely with speed, resulting in an approximately constant com mutation angle T c over a wide speed range. At very low speed, however, this is no longer true because the resistive voltage component consumes an increasing portion of the available voltage time area; the minimal speed, where natural commutation ce ases to func tion is usually below 10% of nominal speed. Operation in this speed range is still possible when reverting to intermittent link current; this is achieved by commanding zero link current, to which the current controller responds by temporarily shifting the firing pulses of the line-side converter to the commu tation limit, i.e. applying maximum negative volt age to the DC-link. As soon as the current in the stator windings has decayed to zero and the recovery time for the thyristors has elapsed, the next thyristor-pair of the machine-side converter is made conducting and the link current can be restored. The delay necessary for reducing and building up the link current can be considerably reduced if a shunt thyristor in parallel with the link inductor, Fig. 14.18, is fired at the time, when the current reference is blanked [LI]; this short-circuits the link inductor and allows the current to circulate freely without affecting the commutation; as soon as the machine-side converter is switched to the new conducting state and the line-side converter is reacti vated, the shunting thyristor is quickly blocked by the volt ages in the DC link. This allows a substantial reduction of the blanking time intervals in the low sp ced region and reduces Lhe effccts of pulsating torque. Tt wn.s JlI(~nt.io!l(~d above ill COJ1lH ~ d,iol1 with Pi!!:. 14.20 Lhat the machine sidc cOllvnl.('t', oIH'.rat.illg wil.lt lIai.llr;d ('OIlIIlIIIt.al.iol1, illvaria,hly C;\.IIS(~S a c(~r Laill ll.lll"lI I ii, "r ('111 ' 1"'111. iII tI)(''' !lxi::, I ,> 1I11;lIl.t>IILillllalloa,d dqwlld.>1I1. :1.1'111;1
:154
14. Variable Frequency Synchronous Motor Drives
14.3 Synchronous Motor with Load-commutated Inverter
Lme reaction. This can be offset by a corresponding change of field current. Olle possibility is to control the magnitude of the induced motor voltages es 1.0 a reference value rising with speedj when limiting the voltage reference above base speed, field weakening is initiated.
Microcomputer 8085 Minimum current
_1_
Supp/y
Converter and motor Link current control/er
~
Firing circuit Line-side converter
Machine-side converter
355
link current reference begins to respond to increased torque demand, The in verter limit angle Qmax must be continuously adjusted to achieve minimal but safe extinction timej it can be determined either by open loop computation based on current, voltage and speed or by closed loop control, possibly with feed-forward compensation from the current. The control loop for the field voltage Up is included in Fig. 14.21 which corresponds to the low impedance rotor model in Fig. 14.13. AIso, control of the field voltage responds more favourably when transient currents are induced in the field winding when load is applied. A control loop for the field current ip would tend to tem porarily reduce the field voltage at a time, when it should be increased. On the other hand, control of the field current would eliminate the effect of changing field winding resistance due to temperature variations and improve the dynamic response in the field weakening region. The sequence of the firing pulses for the machine-side converter can be based on the estimated angular position of the flux model, Fig. 14.16, or on the rotor position ê, if a voltage model according to Eq. (14.33) is used, whose accuracy is questionable at low speed, An important feature is the starting control algorithm in the very low speed range by temporarily blanking the link current via the line-side con verter and firing the shunt thyristor. This is a convenient way of quickly altering the mode of operation, such as torque reversal at low speed which could take too long when waiting for the next natural commutation.
n 1/min
Jo'i~.
14.21. Microcomputer based speed control of LeI synchronous motor drive
wil.h I)C link converter
A control scheme incorporating these features is shown in Fig. 14.21. The cil'cuit contains four converters of different types and sizes.
p()w(~r
• • • •
Intermittent Link current
1 IIII,II~" 0;5
t:he refercncc values for the De link current and field voltage as wdJ II:·; :;i,~lIals for lhe digital firing circuit of the machine-side cOllvertcr a.nel tIl(' lidllllll.illg I,hyrislol', 'l'he function gmleralor 1)t'(~scl'Íhillg lhe Jillk cnlTcnt rd'e r
()r a:;snl'ing cont.iulIOtlH ClllT(~llt;
w!.ic 'h c.ptl ,' hl.v ll !l~H II" ( 'I :
i..Q
20
;1.1'('
1IlOdc' ('(,"!.r()11Ül d lh'l.('d t.llrOIlI~h 1./1 0 Jirillf~ I lll! ';IC~
,/
~
40
f'lI' which the microcomputer is producing coordinated control comman(h;.
('II«T Iaa:; (.llI' pllrpO:j(;
o
-200
- 400
- 600
- 800
A
Linc-sid e converter, Machinc-side converter, ShUllt thyristor, J,'idd power supply,
'l'IH'y
800 600 400 200
iu
t\w lllillÍlllal
c·.tlrrell!.
(,r II.ilc' HlII.dlÍll C:~H iel(' C( 'IlV (', ['i,(' 1
il tl IIppC' 1' III' l"wC'r 111111 ", "
(I CW
(\1
n"'ilHI I,IJ
II H
1.1.,.
1;0
1;5
2;0
11111
,"
2;5
3;0
3;5
4;0
1 5
Fig. 14.22. Reversing transient of a 20 kW synchronous motor drive at half nominal speed
/I. :W kW syllchrouous Illotor drive wit.h microcomputer control has heen c\c'si ).{lIrt! iii UI(' lahonüol'Y aliei t.(~sl.(~d l1un,lu7,IUH1· Fjr;lln ~ 11.22 ckpic:t.s a n·nll,I,·,llc ' vc ·I'nÍtl g 1,1'1I.1I: 1'iC'llt.nt iraI!' IICIIIJillll.l :iJltTCI wil.lr 1.1iC' IIlC'J'tiacl!' LliC' clriv\'
14. Variable Frequency Synchronous Motor Drives
356
increased by a dynamometer without load. The intermittent link current at low speed as well as the upper and lower current limits are clearly visible. The oscillogram demonstrates that the dynamic response is slower than could be expected with a field orientated type of controI. However, there are many applications where fast response is not a primary concern.
14.3 Synchronous Motor with Load-commutated Inverter
357
UTh
-V 200 Thyristor voltage
O -200
n min
-1
- 400
2000
Speed
aO
1500
1600 1000
/
1400
io
Load applíed 1/
A 50
40
30
20
10
O
:
delayed firing causes commutation failure
1200
De Iínk current
ia
1 1 1
A 60
I
I
De Iink current 40
I,
A
20
Field current
5
O
4
b
3
20
30
40
50
t
ms
2
Fig. 14.24. Intentional commutation failure Df machine-side converter
16;~
1400
Firing angle of machine-side ._ __ converteI
1200
~
ms
2
Extinction time
~Ii
10
I
oI
2,0 s
1,0
Fig. 14.23. Loading transient of a 20 kW synchronous motor drive
Further details are seen in Fig. 14.23 where load torque is applied; tlw rise of the field current is initiated by a transient and then maintained by th( ~ controlloop for the induced motor voltages es, thus neutralising the effeds of armature reaction. The trace of the extinction time indicates the good perfonnancc of Lhis crit.ical cont.rol function. Fi II ;tlly) I"i,,;. 1·1.21\ shows I.h(~ dr('cl. or a dd i I>cm tdy l.rif(I~~ md COI II III II t nt.joll L,illll"
wlli"11
, ' :111::1': : :t
1'1I1'id )'i:iI'
(Ir
UI (' lill l\ 1'11111'1'11.. IloWI'VI')', t.lw 1"111'1'1'111.
controller is able to quickly restore the link current so that the machiuc.. :-;idc inverter can resume firing in the regular pattern. As seen in Table 11.1 synchronous motor LCI-drives are being used IIp 1.0 the highest power ratings, mainly for compressors, gas turbines and j.>Ullllwd storage sets. Since the usual method of reducing resistive los ses by illc n:aHill l'. the motor volt ages cannot be applied indefinitely because of the windiJlJ-( iH,' lation of the motor (the usuallimit is around 10 kV), converters iII pnm lll' I connection are often chosen, as is shown in Fig. 14.25 for 12-puls(~ o»(:nl.l.Í()lI , at the sarne time reducing torque pulsations. For this purpose the llIol.o)' wlI.I, solid steel rotor is built with two separate stator windings, displaccd 11.1' :111 " electrically, which are fed from two separate DC link converters. At (.]1(' li ll!' side, a corresponding phase shift may be produced by a starjdelta colllu'dl'cI transformeI. Brushless excitation of the motor with an AC exciter and rol.al. ing rectificrs, common on large turbogencrators, is also indicated. Pi JJ;. II\ .Zfi demon stratcs the cOlltrol scheme, wl)( ~ f( : 1.111: S]lCCC\ controller sUJlplics (:qll:d C1JIT(~1l1. n :f"J'('llc( ';: to t.h(~ s(~pllrat( ~ lillk nlJT('Ut. ClwLrolkr:-; ; tll(~ hac\c voll.a.I'.' ·: ·/I./ll , ·(t/l '.) :wl,iv" iii 1.111" I)C lillk1:i , :\'1'" }1 lJiJ'I,l'd a'" I'<))"(liJl, ~ 1.0 til\! /;I;colI\l"l.l'i l'il.l 1)()h il.i()II:~
"r ii,,·
111"1.,,1' willdill,'. ~ .
1
14.3 Synchronous Motor with Load-commutated Inverter
14. Variable Frequency Synchronous Motor Drives
358
359
iSI(t)
Stator winding 1 Machine-side
converter
Une-side converter
i02
fo= const
.---
fI
=variable iS13
:_~~ i01 (1 '".
a 21
\ I U01
a" iRI
~--f!'~:-'
Rotor winding (combined field and damper winding)
n/6 r-----~----~~--------_,
mM'w JR(t)
Brushless excitation
with rotating rectifiers
Stator winding 2
6-phase motor
Fig. 14.25. High power LeI-drive with 6-phase motor
Fig. 14.27. Simplified winding scheme of a two pole, 6-pulse synchronous motor
Start wilh intermitlent
cu"enl
(1+ aR!LRe
LC;1
I~SI (I)Hül
---.
RS
1@SI
..:....H.J-I.-=l {S2
í
I
ú)
je
mR
(1+ aR) iR e ii e- rr/6)
The arrangement of the windings is demonstrated in Fig. 14.27 in sim plified form; whereas in the rotor a symmetrical AC winding serves as a combined field- and damper- winding, there are two electrically isolated and by 7r/6 = 30° displaced two pole AC windings in the stator; in accordance with the simplifications introduced in Sect . 10.1 they are modelled as airgap windings. The pertinent ampereturns waves are, as in Eq. 10.2,
8 S1 (0, t) = Ns[isl1(t)t9s(o) + iS12(t)t9S(0 -,) + iS13(t)t9S(0 - 2,)],
+ (14.34)
!!S2 RS(I 'O)L S
DI
8s 2(o, t)
= N S[iS21 (t)t9s (o + iS 23 (t)t9s(o -
Fig. 14.26. Equivalent circuit and control of a 12-pulse LeI-drive
7r /6) + iS 22 (t)t9 s (o - 7r /6 - ,) 7r/6 - 2,)],
+ (14.35)
whcl'e - 1 < 79 s (0) < +1 are dimensionlcss periodic functions describiug tJw disiriblltiolls lh(~ windiJlgs. Tw() (·()\J1pl(·.X' Ktntor cllrreut. VC clOI'H aJ'(~ ddill(~d ror t;h(~ t.wo Willdirl1';s,
()r
360
14. Variab!e Frequency Synchronous Motor Drives
isdt) = isdt) e j(,(t) = iS ll (t) + is 12 (t) ej-y
14.3 Synchronous Motor with Load-commutated Inverter
+ iS 13 (t) ej2 -y
(14.36)
= is 21 (t) + iS 22(t) eh + isdt) ej2 -y
(14.37)
= is1a(t)
361
+ jiS1b(t)
and
i S2 (t) = is 2(t) e j(2(t) = is 2a (t)
/
+ jiS 2b(t) .
= N s /2[i s1 (t)e- j'" +is1(t)*e j"') = NsRe [i S1 (t)e- j",]
iS li
/
'"
iSd
" iS23
6 S2 (a, t) = Ns/2 [(iS2(t)ejll'/6)e-j'" = NsRe [iS2(t)ej7r/6e-j"')
+ (iS2(t)ej7r/6)*ej"')
iS2q (14.39)
is2d
iS21
- ((2(t) +11)6)),
-j(p- n/6)
which are superimposed in the airgap,
6s(a, t) = 6 s 1(a, t) + 6 s2 (a, t) = Ns[isdt)cos(a - (l(t))
isld
(14.38)
= Nsis1(t)cos[a - (1 (t)),
= Nsis2(t)cos[a
r-E
isq
e - jp
When including the conditions for isolated neutral points and neglecting spa tial harmonics, i.e. with '!9 s (a) = cosa, two travelling waves result
6sda,t)
iSIQ
iS13
1t/6
+ iS2(t)cos(a - ((2(t) + 11'/6))). (14.40)
Fig. 14.28. Flux mode! of a 6-phase synchronous motor
A corresponding definition holds for the rotor with the mechanical angle é(t),
6 R (a,é, t) = NRRe [iR(t)ejõe- j"'] = NRiR(t)cos[a - ç(t) - é(t)],
.'E".Sl (t) = Lo (14.41)
.'E".S2(t) = Lo
= iR(t)e jW )
(14.42)
is the vector of the rotor current. Linear superposition of the ampereturns produces, assuming the leakage factor 0-5, a radial fiux density wave Bs at t.he stator side of the airgap with length h Ds ((X, é, t)
111 1'
,,,,, ,.1 1.1""
(14.45)
resulting in the flux model in Fig. 14.28. The pertinent voltage equations are .
MoSl (t) = RS!Sl
= (JLo/2h)[(1 + 0-5) (651 (a, t) + 6s 2(a, t)) + 6 R( a, é, t)) .
ftu x ca.UJlot he accuratcly lllodelled with the simplificatiowi ap
plicd ; LIli' t.wo :;(,;\!,or Willdill~~ an~ assullH' d t.o Ilc liJ'lllJy cOllpled huI: tlti:; 'i:i
Wil.h !til" "t'/iIl JI. itl lll . 111 S,'c l.. 10.1 LI", Jlil H VI·rl.OIi I til' 1,111' I.wo :·;I.aLI '" Willdilll',H
[(1 + o-s)(iS1 e- j7r / 6 + 1.52 ) + iRe j (õ-7r/6)]
= .'E".Sl (t) e- j7r / 6
- R'
(14A:-\) Th( ~ It:akag( ~
(14.44)
and
where
iR(t)
[(1 + 0-5 )(is 1 + i.s2 ej7r /6) + iRei''' ]
-
'S!Sl
U ",í,C; "j
d
+ dt.'E".Sl
. . Cj7r /6 ] + L 5 dtd [!Sl + lS2 + L odtd [.!n eje]
(14.4G)
aJld 'I/ ,o.;~ (I,)
d
'I' ,';', .' :.'. '
0[,
(ltl,H)
di, "" f,,· ti " di
I
l 'II "
, n(1l I
; ,I~
II
ti "" ,1/
I
1,..' ,I( .
,, / 11 )
I,
:\62
14. Variable Frequency Synchronous Motor Drives
15. Some Applications of Controlled Electrical Drives
The rotor voltage equation is
UF(t)
d
= RRiR(t) + dt'KR = RRi R +
Lo! [(1 +
G"R)i.R
+ (i Sl + iS2ej7r/6)e-je] ,
(14.48)
where UF(t) is again an assumed impressed volt age source serving for DC ('xcitation. The design of the control corresponds to that shown with the G-pulse drive. The preceding chapters were dealing mainIy with the different types of eIectri cal drives and their control; applications were only mentioned as they affected the operation of the machines and the associated equipment. AIso, the driv ing specifications of a mechanicalIoad are normally not met by just one type of eIectrical drive and the variety of appIications can be bewildering; in this chapter some problems associated with controUed eIectrical drives will be ex pIained in more detail. For this we begin with a 4-quadrant drive, be it DC or AC, the basic structure of which is contained within the dashed lines in Fig. 15.1. The moving masses are at first assumed to be rigidIy coupled, forming a Iumped inertia. The inner Ioop which comprises the power converter and part of the eIectrical machine assures fast torque control; with an integrating controUer it exhibits unity gain and serves for linearisation. Once the torque Ioop with the equivaIent Iag Te is closed, there is littIe difference between a DC and an AC drive.
mL
-_._-----,
L____________________________ lJ
Fig_ 15.1. Spced controlloop with rate-of-change-limiter and load feed-forward
'l'orqllc ("olll,ml is Illiwdatol'y on hi gh pCl"formance drives except for the :-o l\i..!I.. :: 1. pnwnr f ll.l, I II I{H, "<:("'IWS\ ~ I:nrqll(' :wrVI'ii a..' i I.\\(~ controlling input t.o any 11I1 ·I'h II.II II·n l ,,1 11111.. WI'''liI"V''r IWI,II Lo ... " ... IIJll nl, 1)(' l 'Olllll.l·l 'iV·j,I·d 01' 1.11 0 s p p,..d
:\(i~
15.1 Speed Controlled Drives
15 . Some Applications or Controlled Electrical Drives
is \.0 be changed, it is only possible through the torque reference; hence the J"('sp onse of the torque controlloop limits the control bandwidth of the com I ,I(,\.e drive. ln view of the practical difficulties of measuring mechanical torque over a wide frequency range it is most desirable to have an electrical substi 1.1I1.(~ available as feedback signal; since torque control is usually embedded iii higher levei controls offering a corrective input, steady state precision of I.lw t.orque signal is not a primary concern as long as the dynamic behaviour Dr I.h e real motor torque is properly reflected. By limiting the torque refer ('IICC, protection of the power converter, the motor and the mechanicalload is achieved. Sometimes it is possible to extract an estimate of the load torque, mL , Irolll a load model and feed it to the input of the torque controller as a dis t.urbance feed-forward. This is called inverse load modelling; it results in t.hc best possible dynamic response of the drive that could only be improved wiLh a faster actuator, i.e. an increased control band-width. For the following the closed torque loop is considered to be part of the driv (~ equipment. For safety reasons it is normally not accessible to the user I)('Ciw se its internal functions may be too involved and subject to critical ;u l.iu stments.
365
To avoid the transmission of large fluctuating torques through the shaft which would result in undesirable torsional motions, each section is driven by an individual motor that is supposed to carry most of the sectionalload so that only small synchronising torques are transmitted through the drive shaft; the load in the different sections and, correspondingly, the motor sizes may be dif ferent. Similar problems arise with multiple motor drives on paper machines of earlier design having a mechanical drive shaft or on trains with multipie drives . Load n
~
'V~---.J)(
a
Section n
Load n +1
v,------'
Section n +1
15.1 Speed Controlled Drives Ily slIpcrimposing to the torque control a speed controlloop the most fre qll( 'lll.Iy lIscd drive control scheme results; its steady state characteristics areS dr;l.w lI iII fi g. 2.6, where the torque limit may be adjusted to twice nominal I.m'lll(' . Wh en the speed reference WRef is manually set, it is normally specified "II 1;lrl\er drives that speed changes should take place with less than maxi 11111111 l.orqllC; on the other hand the control should respond to load changes as 'illicld.v as possible and with full torque. These at first sight contradictory de 111;l.llds are met by making the speed loop as fast as is practical but including ;1. r;d,(,-of-change limiter at the reference input, outside the controlloop. This i.'; .~<:Cll in Fig. 7.11, also showing transients of the speed reference following a ~; I . cJl chauge of the speed target-point. The rate-of-change limiter may be in I.('rprd.(~d as a simple dynamic model having a desirable and safe response, no III;d. l.t~r how quickly an impatient operator may be changing the speed target. SI)('('!.or d l'i v(' 011 ro\.ary 11I' ill\.il q( prt'; IN":'t , wlwJ'(' ('ip:h\. or JllOn' i II . (1, 1'1 <1 11 ' ,[ 1"1111.11'1 ', 11 1,1I 1.1" II HIli fW 1,(' II ll'dlltlli(·l\ ll .v "oll"l, 'd I,'y II. 1"111 ', dl'l vp 11 1I :d'l"
b Fig. 15.2. Multiple motor drive with mechanical drive shaft (a) Mechanical systemj (b) Block diagram
Since the motors in Fig, 15.2 a are semi-rigidly coupled, only one speed controller is needed for which the feedback signal is taken from a suitably placed speed sensor. (Bach motor is usually equipped with a tachometer because the drive system may have to be rearranged by opening and closing the clutches.) Sharing of the total torque is achieved by the torque controllers which receive their shares (k n ) of the total torque reference from the common s pced controller; a simplified block diagram of this control scheme is drawn iu Fig. 15.2 b. WhclI tll(' 1I1Ot.or~ of a multiple drive system are only loosely connected, II. dilr(~l'clIl. ~i t.lI;d . i()1\ (~XiH t s , whcre eadl lJIotor Il( ~(~ds its own speed controller. 'I'hi ~1 1'1',,1>11'111 ilri:«':; r"r ill lil.a. Ilt'(' ou larl~" IHLJ)( 'f" llla.chiJl c:; withollt nWrhaJ l~ t'l d d il"" :illid·1. " xl."11I1 1111'. ",," I II 1"111:111 .. /1 I()(] III 0 .11,( ""IlI.lJ.illilll~ d"'{,"lI li "r
366
15. Some Applications of Controlled Electrical Drives
15.1 Speed Controlled Drives
individual drive motors; similar structures are found in continuous hot strip rolling mills and many production processes in the fibre and textile industry.
a
each motor must therefore be maintained in precise relative synchronism with the preceding and subsequent sections (ignoring constant or slowly varying scale factors such as gear ratios or rolI diameters), The speed ratios must be accurately maintained independent of the speed leveI that may be changing during the process and with the type of production [K28]. A proven strategy for this type of drive system is to derive the speed ref erences in a serial form, so that the ratios of reference speeds in neighbouring sections, n
Qn OOn
V ReI
OOn
OOn+1
• Better accuracy, • Possibility of storing reference data and • Ease of recording and evaluating measurements.
c Fig. 15.3. Sectional drive system for continuous production process (a) Mechanical systemi (b) Block diagram wi th serial reference chaini (c) Block diagram with parallel reference structure
A characteristic of such drives is that the individual sections are only lightly coupled through a strip of red-hot steel or a continuous web of thin material that cannot sustain any appreciable forces; on the other hand, th( ~ material must be kept sufficiently taut to avoid creases or folds. Fig. 15.:l shows a schematic example of a continuous process with a multiple drive ::Iys !.PIII. The velocity of the material is usualIy increasing as it procecds throllgh UI(' diIT(~n~lIt slatiolls; this is ohvioll~ wiih the continuOIIH rolliug mill, whcn' 1.111" 1J1I't,:d :,hl'\'l 1!('COIL)t':; t,hill\l(~r alld IOlllSl H' wlll'lI p n.... H illl~ t.lu'ol1lo(h ('a.dl pair 01"
I'II11 rl 11111, H.pplil'!I ,'q\lfl.lly l(,
01\ '1\'111 1>1 1'1'
""I.y I...
11 I'lI u ' I,i"lI
II
"I'
Imp"/' IIl1lChil\(" "III' P"'I " ' !I!,
III
('Vl'1I
nlll'lI
I
W Wn+l Ref
remain constant even if the speed ratio may have been changed in another section of the drive. This control principIe, calIed "progressive draw" is in dicated in Fig. 15.3 b; a change of the overalI velocity reference VRef or of a preceding ratio simultaneously affects all folIowing drives in the desired way. The idea of achieving "progressive draw" by deriving the speed reference W n Ref from the measured speed wn+l of the preceding drive can be dismissed because after a few repetitions this would result in poorly damped and even unstable transients, similar to a queue of motor cars where the attention of each driver is solely fixed on the rear of the leading vehicle. If all the speed references W n Ref were derived in parallel from the com mon velocity reference VRef, a change of one speed ratio would require alI subsequent settings to be changed as welI; this scheme which has merits in other applications is sketched in Fig. 15.3 c. The control variables are usually processed with analogue means including operational amplifiers or precision potentiometers but digital methods are becoming more widely used; the main reasons are
OOn.,.1
b
367
t,holll (h t,lt" HI){','d iII lH'f !.i ll II ,
'1'11 "
11 1"'1'"
"I'
The improved accuracy of digital methods is due to the fact that spccd:; can be measured to high resolution by counting angular increments and tha.L very precise reference frequencies are readily available from crystal oscillatorH (.dI I lo < 10- 6 ); also digital controllers exhibit no drift effects. An exampk of a digital speed control is described in Sect. 15.3; typicalIy, the specified error of speed ratios on a large paper machine producing 10 m wide paper at a velocity of 30 m/s may be < 1O- 4 /day, which would be very difficult to attain with analogue methods. As wprcciabl(~ <'la':i t,iril.,Y, 'I'lti~; ,'ollld I", tl\H~ 1.0 rl loll g dl'i V( \ Hhllfl, or 1(('iU'H s eparatill ~ I.)w lllotor rI'O II I 1,111' 1"lI tI; 1.111' /11 1.1111' iH I.rul' /'01' ('1' 1111\ '11 "' lt" i ll l. H WIt"I'\' t.11I' ('!aJI I,i, 'il,,Y 01' 1.1,.. ,'''''\' Íil I'''!. "1'1',111',0101" 1,1111, 1t'.~ I . 11",,1.111'1 I' XI II"I" " III " 1"1,,,t.i! , wlll'lI ' ,'ln :: U.. II,,Y
368
15. Some Applications of Controlled Electrical Drives
15.1 Speed Controlled Drives
is the consequence of a lightweight construction. A model of this type of drive is depicted in Fig. 15.4 together with a simplified linear block diagram. K is the torsional spring constant of the ftexible transmission; its inertia as well as its inherent damping are neglected, ê1 -ê2 is the torsional displacement of the shaft. The load torque may again consist of various components induding lift torque and friction, it is represented here by a variable independent torque mL. ln practice part of the load torque would be speed-dependent and, hence, could exert a damping effect.
me
mM"úl),E)
a
J2
A
,-' K ',.---------------------
_---I
'.. . _/
(~''l.!!~'!'2t~!...sE:.e!~e!'!.b!:k
: :
Speed I controller :
/
:
:úl) - úl2
:
Torque control toop
I
I
l...---:-l
I
mL b
~
/ I',
.0
j~~:i ® j~
= J 1 wo
Fig. 15.4. Two-mass drive system with torsional elasticity (a) Mechanical scheme, (b) Block diagram, (c) Bode diagram
The inner torque controlloop of the motor, which in the case of a De drive wou!d be the armature current loop, is represented by an equiva!enL !ag !mvillg unity ga.in and time constant Te; it must react sufficiently fn-'il. Lo dfccti vdy mntrol this type of plant. The reason is again that eorrcctiv(! a.ctioll (·.all ollly bc il]lplied aL the refercuce sidc of the torq\l(~ coutrolkr, i.('. I.hnJUI~h I.h<: I.orqll('. ('.out.rolloop. Tltc o:;c.illat,ory lIwdU\.\Ii(~ftl par\; of t.J!
wit.h t.wo illl.('J':raLol'!: ,·,u·h. TIII!
I,wo illll"r
I'IIU ('l.[,,1I
1... I.W''I'fl Lor,!,/(' lLIHI
IOO(l!i
1110(.01'
!i lll·,'d i:l,
mo
T 2 = hwo
'
mo '
il01
=
K
(~+~). J1 J2
il02 =
I!f. (15.2)
2
F2 (
c
,·olll .l l,ill !i
(15.1)
OIH'!I
-
T1 T 2
»
1
i.e. if poles and zeros are dose together and are nearly cancelling. This is seen in Fig. 15.5, where measured step responses of the speed control loop are shown for different cases following changes of the speed reference W1 Ref and of the load torque mL. The transients were recorded on a DC drive, with the resonant load being simulated by another controlled DC drive. The parameters of the Pl-controller were in each case so chosen as to obtain optimal results [B72,B73]. It is realised that, even when the motor speed follows the reference speed in a swift and well damped transient, the load speed, being coupled to the motor speed through the transfer function
me
IFI
T1
J
E) - E2
: :
, - - - , O) 1
1 (8Iil o2 )2+1 (T1 +T2 )s (8Iilo1 )2+1'
with the following abbreviations
_ J1 _
:
/,,_,
I 1_ .------------.
L (wt/wo)
= L (mM Imo)
It contains imaginary pairs of poles and zeros which make this a diflicult plant to control; the Bode-diagram is sketched in Fig. 15.4 c. When attempting to control the motor speed W1, which is the usual prac tice because the speed sensor is normally attached to the free end of the motor shaft, it is realised that a reasonably damped speed loop with a Pl-controller can only be achieved under the condition
úl2,E2
~~)mL ) me me:
F 1(8)
369
Joop
I.rall! if'·1
8
)
L (w2Iwo) _ _ _1 - (8 I il02 )2
= L (wt/wo)
+1
(15.3)
remains outside the control loop and can exhibit practically undamped os cillations. Obviously, this calIs for a different approach to the control of the planto When assuming for a moment that in an ideal case both speeds W1, W2 as well as the torsional displacement ê1 - ê2 of the shaft are measurable, then a very effective scheme for decoupling the mechanical plant could be devised, which i~ indicated by the dashed signal paths in Fig. 15.4 b: By feeding back to thc torqnc controller a. signal whirh is proportional to differential ::ipccd, dalllpiJlg is achicved h(!CiWS(' tJl!~ two 11l;\S:';(!S are kept aligncd, thu::i r(!llJovil1l~ t.Il!! f:allN(' ror :mhSC
15.1 Speed Control\ed Drives
15. Some Applications of Controlled Electrical Drives
:170
J1
mm-'~ ~
II
200
= 10
J2
L:=
ro1
-20~P
mln-'
200
~
measures is that the torque loop is responding sufficiently fast, Te « Tt, T 2 . The effectiveness of L1w-feedback for the speed control ofthe two mass system is apparent from the recorded transients in Fig. 15.6. UnfortunateIy, feedback signaIs for speed and position of the load, W2 and C2, are usually not available. There may be practical reasons, for example that the load is driven though a gear and rotates at a very Iow speed so that mounting a speed sensor would be difficult and expensive. Similar problems exist if the motion of the load inertia is translationaI, for instance on an elevator or mine hoist.
J l = 0.2
m,
Jtvv\
J
2
fvv:=
t -20~ ~
Step change of speed reference
m;~t~
~ m~~ ~ t_20~J P ·
P
_20~J
j
---J'
~ rol
ro1 --..
m"0:::1))) 00,." m'f) T)J m
f\-...
t
L
me K
Microcomputer --i-Drive I
v"-...
t
371
ContraI plant
Step change of load torque
1~t
a
b
~m2
v
me
,,_
\.--õ+--mL
Fig. 15.5. Transients of speed control loop with two-mass system, recorded with :. 1\0 kW DC drive (a) J 1 = 10 J2 ; (b) J1 = 0.2 h
II
~.L
Drive
__________ _ ----------ControT:--------- .. ------,
"Load feed-forward ~ú)-
~ú)-
Feedback
P
of(
of(
on
. -1 /1)/1.' 200 .
.o~p Jl J 2
_
10
2s
m;nl~ 200
' 20~
DamPing~I
I
t=
11
on
~
t
min-~ I 200 o
-200
1
,-
roi-roi 1
r----- 4
-'+
-'+ I I
II
Feedback
L
~
~
t
J 'f = 0.2 2
t:=== min-'~ 200 ~
\;--J t -20~1P ·
Fig. 15.6. Effect of auxiliary Llw-feedback on speed control of a two-mass system, recorded with a 40 kW DC drive
'-... Input to model
®
II
Output ------ ofmodel
I~ I [email protected]~_~i-i
rol ___
-
- ro
II
k
2
I~
...---+-I
k3
I
1 1
~e
i
i I
1
11
II
II
l_ rol_- ro?________________ ....!
~~
(Observer)
Pig. 15.7. Speed control of a two-mass drive system with auxiliary feedback signals derived from an observer
ln such a situation one could attempt to derive the necessary information from a dynamic model, called an observer [W17, W18], which is shown in Fig. 15.7 in simplified formo Thc observ(~r is a mathematical model of the plant tllat is (~valllated in. real time iII pamlld witb the actual transients of the pJ a.1I 1.. 1':Kl.ablishiJlg thiR modd 1'('qui"" 1"! I.hf~ s t.l"IIdlll'( ~ Rlld thc parnlllc krs or 1.11(' plalll, t.o Iw knowlI fnirly ILI" 'III'il,I,,·ly, wllich iK IIsllal1 y uo I.~n ' at. prohl"11I wll.h 1I1""hHlli(",d Ky fi t."III IL Til, · Iii " !'I V"II hy t,lu' 1)1( ~ I I.H llrll.hl('
III""'"
i IlPIlI,(I "I' 1.1", ,,111111.,
'III
I.h ln
C / Iii"
1,lI u 1."1 ''1''''
1" '1"1' 1'.1 11' "
wh1rh iII ,dl'Pfu l.v IWII,II
15.2 Linear Position ControI
15. Some Applications of Controlled Electrical Drives
372
able in the microcomputer; the remaining inputs, particularly the load and frictional torques, are unknown and have to be estimated. Keeping the model in line with the actual plant is achieved by comparing some accessible output variable from the plant, for example the motor speed Wl, with its estimated counterpart Wl and correcting the model by additional inputs so that the er ror vanishes. If the model parameters are correctly representing the plant, it can be assumed that the remaining estimated variables in the model are also tracking the real variables, i.e. that the model serves as an accurate observer. Clearly, designing an observer calls for a detailed analysis of the system to be modelled [L80]. The structure of the observer seen in Fig. 15.7 gives an idea of how the correction of the model can be achieved; each of the integrators in the model is influenced by the error Wl - Wl in order to obtain maximum flexibility in selecting the eigenvalues of the observer by suitable choice of the gain factors k i . This is so because the observer represents itself a feedback system, the clynamic properties of which are the subject of a separate analysis. Realisation of observers for complex applications has only become prac tical since distributed digital signal processing with microelectronic compo lIents is possible; neither analogue techniques nor a main-franle process com puter could provi de an economic solution.
Ú)1
Ú)1
ú)o
ú)o
0.2
0.2
0.1
0.1
373
rather unsatisfactory transient considerably by applying differential speed feedback and damping signals from the observer; the oscillations are now all but eliminated. The algorithms for the speed controller and the observer have been implemented in a single board 8086-microcomputer; the sampling period for the complete algorithm was 2 ms, the storage requirement < 1 k byte of ROM. The characteristics of the plant and its parameters must be known for the computed variables produced by the observer to be trustworthy. While the dynamic structure of mechanical systems is usually known from the ge ometry, the distribution of masses with their linkages etc. may not be known accurately or may change under varying operating conditions. EXanlples are mine hoists, where the winding rope at different length acts like a variable spring, or robots with changing geometry when positioning a mass by a ro tary movement at a varying radius. li the parameter changes are known or can be derived from other sources of information, it is of course possible to include them in the model, forming a time varying observer; this is feasible if the observer is implemented in a microcomputer. However, if the variations of plant parameters are unknown or even the structure of the plant is uncertain, the problem of estimating internal vari ables becomes much more complex because it involves the identification of the plant as well as the adaptation of the observer and the controllers. This task has also come now within reach of a microcomputer but it would require a more detailed presentation than is possible in the present con text [B73, M21, WI3].
15.2 Linear Position ControI
O
a
2
t /s
0 1
b
2
t /s
Fig. 15.8. Step response of speed control of a two-mass system, recorded with a I kW DC drive. (11) PI-speed controller; (b) P-speed controller and observer
Tlte cstimate of the load torque, mL, is generated by integrating the Sl)(-'(~d (~rror WI - Wl; this permits the use of a proportional instead of a proportiollal~ pllls-illl/~g;ntl (PI) speed controller thus reducing the speed overshoot followiu ii, cllallg(~ o[ lhe rderence without sacrificing steady statc accuracy. Ileslll t.H ror a I kW De drive wit.h microcompntcr cOlllrol af(~ shown in Fig. lG.H \ W I ~ , W 1:1\. '1'1'1(' :iI,cp r<~SpO)lSC of lhe sp(~(~d ("out.rol loo!> wit.h nu opt.irllll.lly n d J lI l-l tt'd \.'1 11 1)(,,'·d ,:"ld,l'olll
ln many applications of electrical drives it is specified that the shaft of the motor or the load should be controlled to a constant or variable reference angle or that a machine part, possibly the tool of a milling machine, should follow a prescribed trajectory; this rotational or translational movement may be part of a multi-dimensional motion. It is generalised on a robot having six or more position-controlled drives to be able to place the tool in any position of the workspace and point it in any direction. While position-controlled elec trical servo drives at lower power rating (usually < 10 kW) are of particular importance as feed drives on machine tools or robots, this is by no means the sole area of application. Servo drives are found wherever mechanical motion is required for controlling technical eqllipment in industry or transportation, be it the püsitioning of servo-ac:tuated valves in c:hernical plants and power stat.iOllS or tI[(: ddlcct.ion of c:olltrol surra!"('s OH ali aircrafl. lndeed, position ("ollt.ro[ i,·; al.';tl rtllllld 011 \Jjg[l(~r I)(JwI~r dl'ivI~H, HIl('.h as dl'w\.tors, mine-hoists or :\.llI.Olllll.l.il' (' 1'.111111111.1' " t.rninH . IkJlI'lltlhq" '"I LIli' H,ppl i l' ll.l,ioll I.ILI'I"
11\'1'
"r
1" 'IIl'1;"
dil!','",,,1.
1I"ll lÍ n'lIlt'''I.:.;
wit.1t l'I 'j', II,tI L.. I H'l' W !Il' Y
III
1.111 · 1',, 1
;\71
15.2 Linear Position Control
15. Some Applications of Controlled Electrical Drives
lowing sectíons. At the sarne tíme they demonstrate again the immense vari ahílíty of electrical drives. Let us assume at first that the posítion reference and the feedback sígnal I'cpresenting the measured position are continuous or sampled variables and l.!Jat a linear position control scheme is desired. Specifications of this type ;tpply for example to elevators, servo drives for radar or satellite antennae or Lo fced drives on machine tools for contour milling. If the reference input is a l!lechanical signal, such as the position of a pick-up sensing the contour of a Illechanical model to be duplicated by the machine tool in a different scale, I.he control system is also called a position follower, but in most cases the position reference and the feedback signals are represented electrically. It is further assumed for the subsequent arguments that the rotation of Lhe motor shaft is converted by an ideal gear into the translational motion of a machine part having the velo city v and the position x. Transmission errors snch as nonlinearity of the lead screw or ambiguity due to backlash can be reduced either by employing a precise spindle or by a direct position measure Illent ínstead of a rotary sensor attached to the motor shaft. Posítion sensors are available in great variety, operating on different principIes and covering a wíde range of accuracyj ín some cases a simple potentiometer or resolver may slIffice, while digital encoders or high resolution incremental sensors may be n~quired in othersj ultimate precision can be assured, if necessary, by opti cal methods employing laser interferometrYj the cost and complexity rise of course steeply with the specified accuracy. The principie of cascaded control was explained in Fig. 7.5, showing a system of several superimposed controlloops for torque, speed and position. 1\ II aeceleration controlloop which has the purpose of eliminating the effects or load torque, is occasionalIy added as an option, usualIy on the basis of an ;I,pproximate feedback signal derived from the speed measurement. Acceler a.l.íoll control is no substitute for torque control, however, because it cannot prcvent static overload, for instance when the drive is mechanicalIy jammed. The advantages of nested controlloops in a cascaded system have been described beforej they are • 'l\'allsparent structure, employing alI measurable state variables • Siep-by-step design, beginning with the innermost loop, thereby solving the stability problem in several smalIer steps, • Us!' of standard controlIers, P, PI, PlD etc. • Tll(~ eJIects of nonlinearities, for instance in the power actuator, are re strict.ccl to the inner loop, resulting in a quasi-linear, unity gain response, • Díst.urbances ac;t.ing on the plant are quickly couteracted by inner loops, • 11I(.(~t'Il1(~diate variables can be limited by clampillg the pertínent referellC<~ val,jal,ks, • COIlUIl ; s~ í()l\illl~ is grcatly silllplificcl hy c!osillg Oll( ~ cOIlt.rol loop anel' t.1\(' Ol.lll"f", fi"I\l\ illiiid,' wlf"
375
• Opening of outer loops permits simple procedures for diagnostic and field tests. At first sight there is only one serious drawback of the cascade control struc ture as seen in Fig. 7.5, that is caused by the fact that the response to the reference input becomes progressively slower as more outer loops are addedj it can be shown that under simplifying assumptions the closed loop equiva lent time constant Te increases by a factor of at least two for each additional loop [K25,38)j hence the response to the reference input of a multiple loop control system may be slower than the response of a corresponding single loop system - provided, of course, that a stable single loop control can be devised. As a consequence of these accumulating delays the position control scheme in Fig. 7.5 may exhibit unacceptable dynamic errors when the reference position is time-varying.
tTarget
Reference generator
(Dynamical model)
'Rer
Aro Rer
Position contral/er
t'.mMRef
mMRef
r~~ ~.I !.. ___ :.J Speed control/er
I I
t I 1 1
Acceleration control/er
~____
(ú
t
la I
I
:
----i--------- J
Sígnal processíng i Drive and mechanícalload Fig. 15.9. Multiple-Ioop position control with feed-forward from a dynamic model of the reference trajectory and the load
Fortunately, this disadvantage can easily be removed by employing feed forward to tlw illll(~r loops, as is shown in Fig. 15.9, where a reference gener :l,l.iI ,ilil.y, "l.Il1'l'wi: :j'
376
15. Some Applications of Controlled Electrical Drives
there could be contradictory demands with the result of saturating individual controllers. On multiple-axes servo drives such as used on continuous milling ma chines, where the tool should accurately follow a given spatial contour, the set of reference signals (x, v, a)Ref is usually computed off-line for all axes and stored in a memory to be fed in parallel to the various drive controllers. Another solution is seen in Fig. 15.9 where the reference variables for each axis are generated on-line by a dynamic model representing the desired response of the drive; the effect of this scheme is that temporary satura tion of controllers is avoided which could occur if unrealistic (for instance (liscontinuous), reference signals were applied; by employing a suitably cho sen reference model the dynamic behaviour of the control system becomes reproducible and dependable because all the controllers remain in their lin ear high-gain mode and keep the respective errors smal!. Naturally provision must be made for unexpected disturbances such as load torque or Une voltage ftuctuations which means that the torque control loop must have sufficient margin for absorbing the additionalload, should it occur. The existence of a dynamic model can also be very useful in other applica tions; for example, controlling a high-speed elevator swiftly and still agreeably for the passengers requires that certain limits of acceleration a(t) and jerk da/dt should not be exceeded; this can be achieved by a dynamic model producing a set of smooth reference variables. Since the feed-forward signals improve the precision of the trajectory of motion, it also becomes possible to itccurately predict the stopping distance given any initial condition (x, v, a)o; this is important for being able to decide instantly whether a call from one of the forward ftoors or a late stop ping demand by one of the passengers r ,m be answered immediately or whether the caller has to wait for the next cItance, as speed reversal between scheduled stops is ruled out. AIso, position (lvcrshoot is not allowed on an elevator because the passengers would dislike it a.nd on a machine tool as it could leave marks on the work piece. The ultimate in dynamic performance is obtained by feeding an estimate of tIte load torque mL directly to the input of the torque control loop as is illdicated in Fig. 15.9, thus relieving the speed controller from generating the lIlain part of the torque reference. The torque estimate must be based on a l!lüdel of the load and the reference trajectory, definitely not on measured vallles of acceleration, speed or position, as this would imply a danger of illstabiJity due to unavoidable inaccuracies and computational delays. Sincc rOl"r!lillg an estimate of the load torque for a given trajectory of motion effcc t.i vdy i llvolves thc inversion of the cquation of motion, this is also called "fccc].· forward hy illV(~rSe load modclling" [F28, L41 J. With multiple axes robots t1w iuvC'l"SC lond rnode] Illay be of considerable complcxity; a simple exarnplc waH S!tOWll iJl I<'i, ,;, 2.11. \lVIi"1I COIl Lrolli "I', I.JJ(~ 1Il0t.iOll 01' a lII;\chilll~ /.00\ Lo hi~It [lfc(:isioll, for " ~ !I'''I/li(' wll.1i ii. lI)il ~ illllllll po~; iLioll ("TOr (li' III I, 'III. or 11, 1.01.;1.1 l.r n v!'l di:;L;I.Il('('
15.2 Linear Position Control
377
(Set)
± ó. x Re' (Incremental)
Position error
& '/.Ü)
±ó.x -
/1'
Incremental
sensor
Fig. 15.10. Digital position control
4 1, Zero mark
7
8
a absolute position sensor
7
O
5 8
11 12 b incremental position sensor
Fig. 15.11. Absolute and incremental position sensors
of 1 m, it is necessary to employ digital signal processing not only for the position reference but also for the feedback signal and the erro r detection, Fig. 15.10. This is the only way to avoid drift effects which would Occur with analogue signal processing; only a digital controller is capable of determining with certainty minute differences between two large numbers, for example 25,000 and 24,999. Since a digital speed controller does not exhibit drift, and position is the true integral of speed, the position controller can be of the proportional type; there is no need for an integrating channel for achieving a high steady state accuracy. Naturally this calls for comparable precision also of the digital sensors for rotary and translational motion, combining high accuracy and resollltion. An cllcodcr, Fig. 15.11 a, prodllcc.:s a ll alJsolute llleasurement of thc dis taIl.C(~ 1'1'0111 ;1. Jix( ~d posilioJl, for CXft1l1p\, ~ hy Hllpplyillg II hit of a dual won\ cOV('rill/'; l.i1" 1';111/',( ' 01' O ::S :/: ,, (:.>." 1) / \ , W\H']"(' L\ i:; I.\lI' 1:(':;01 11 t.iOll. II. is I'J'( ' ("( ' I ill>I,· 1,1) ('/lII'IIl'y ('"d,'s wIi,'I'" olll,'y "II" I,il, dlll.II/'.(':: nl. n.lIy OIW till\(" LI)
JlI
I l l- /\
15. Some Applications of Controlled Electrical Drives
;\78 (~xclude
ambiguity (Gray code); a similar effect is achieved with a synchro Ilising signal from the channel with the highest resolution. Another somewhat simpler sensor is of the incremental type, Fig. 15.11 b, which generates a forward or reverse pulse for every increment of traveI; by counting these pulses, the actual position is obtained, provided the counter ltas been initially set by a separate calibrating pulse or some other accurate !losition measurement. Incremental sensors which normally function on a Illagnetic or optical basis, are sometimes considered unsatisfactory because pulses might be lost, causing a corresponding undetected position error until lhe calibrating position is passed again. However, with todays sensors, using LED light sources and the associated integrated circuitry, incremental sensors can be regarded as very reliable components; if desired, battery-buffering of I.he counter can be provided.
E
=f(xb/ xa } ~
úl
generation Counter
are tan Xb / x a
=M ôe
Microprocessor
tr
'nx slg asignxb
/
~.
15.2 Linear Position Contrai
The function of an incremental sensor for angle measurements is explained with the help of Fig. 15.12. There are two optical or magnetic measuring channels producing out-of-phase signals, approximating X a = x sin N e and Xb = X cos N e, where e is the angle to be measured and N is the number of periods per revolution of the sensor shaft; a typical value is N = 212 , By clipping these signals, two orthogonal square wave functions sign X a and sign Xb are created. Whenever one of these functions is changing sign, an incre mental motion of Lk = ±21r/4N is registered, where the sense of rotation is derived from the direction of the signal change and the value of the other signal. With N = 212 = 4096, this corresponds to an angular resolution of 2- 14 or about 1/16000 of one revolution of the sensor shaft. By analogue interpolation, also shown in Fig. 15,12, this resolution can be greatly improved [845]. The signals X a and Xb are not be exactly sinusoidal and the amplitude x may depend on the speed of rotation w = de/dt, but by forming the ratio Xa/Xb and computing arctanXa/Xb, severallower bits of angular position measurement are generated, increasing the total resolu tion to 18 to 20 bit, or 250 000 to 1 Mio increments per revolution. While the dependability of the additional bits may be somewhat in doubt, the ana logue interpolation serves as a most welcome increase of the measurement bandwidth when detecting slow rotational speeds. With both types of position sensors it is easy to perform a digital velocity measurement. By sampling in short and precise time intervals T the position signal available at the output of the encoder or the counter and subtracting subsequent readings, the average velo city in the last sampling interval is obtained.
II
v(v NE
NE
379
1
+ 1) = T [x(v + 1) -
x(v)] ;
(15.4)
correspondingly, an acceleration signal can be detected from subsequent ve lo city measurements,
I
-1 Analogue sensor signals
I~ - __
I I
. 11eV Incr 1 MIO
2 12 ~ 4096
Incr rrev
Fil{. 1r.. 12. 1':IIIIl1'i'li Jl I~ Lhe f""olut.ioll of ali illcr<:lIlC ~ III. ltl ",II("od<:1"
I).v
f\ludctl',I IC"
illl.' ~ J'p()I l\, till1'
1
+ 1) = T [v(v + 1) -
v(v)]
1
= T2
[x(v
+ 1) -
2x(v)
Digital
sensor signals
-$-~ "O,~
a(v
+ x(v -
1)] .
(IS .!:! ) With a given sampling interval T the resolution of a digital speed measur< ~ ment is reduced at low speed because fewer angular increments are aeen mulated in each period Ti the limit is reached when only a few counts aJ"( ~ providing a very coarse representation of speed with questionable vaIue. Thiii then calIs for either increasing the resolution of the sensor, for instaooé by analogtlP intcrpolation, 01' enlarging tlte sa.mpling period. This could be dOJlI' eitlwr hy dlOOliillf.1: a. IIlultiplc peri od mT, 1.1lUR reducing the bandwidl.h 01' LIt( : llW:LH \ll'(~IIWJlL, d f~ ~ (" rilw",
Lo a
01'
Lili/('-
\)y
eOllvC'.ftilll{ 1'/'11111 n
hn H(:d
Ii i
"""111.".1
ff( ~q\lcll('.y-
bascd
nH~aS llreJllCIlt" ;a r;
wl!('n' I.I/(' iul.crv;a.ls Iwt-w('cll
Hllr
Ili, (Ia fl'l''i"''II('Y l"<'fon'Il('( ' a llcl iIl V'·II.,'c1 rol' ,,1,1 11111 1111'. 1\ 11" .. ,,,1 ili /\ " III , VVil.h I\. II lic ")"J("""1 i'" 1."1' ,
("('HAiv c 1I.1I1 \ "l a l" ,,"I :l ('rlI Uf' Lh o n'llI dl,
Ill('I I.H \Ir'I~ II/(,IIL,
w l l,h IL li l,ro,hl, '
:\i\O
15. Some Applications of Controlled Electrical Drives
rrcquency- and time- based sensing may be carried out in paralIel, selecting wilichever result offers a better resolution. The resolution ofthe speed sensing can also be improved without analogue i Ill.erpolation by forming in each interval the integer difference of the incre Illcut.al counts and simultaneously measuring with a high frequency reference til(' differential time between the first and last increments [K10]. ln the past, when digital signal processing equipment was bulky and ex jl('llsive, it was only used when it was absolutely necessary to achieve the J'1 ~qllired accuracy, i.e. in the outer position controlloop of a servo drive, but wit.h todays microelectronic hardware this is changing rapidly. AlI the control rlluctions in Fig. 15.9, including the control of an induction motor and the !('rcrence model, can now be incorporated as a modular algorithm in a micro cOlllputer which finds room on a postcard size printed cardo If necessary, the s; ullpling time in the order of 1 ms can be reduced by a factor of ten when (,Illploying signal processors processors; hence DC servo motors are rapidly sllpplanted by digitally controlled and compact AC drives. This is a remark ;d)!c development made possible by the joint advances of microelectronics and jlowcr electronics. It was mentioned before that backlash in gears and couplings may cause jlroblems with position controlled drives because the contact between two ;t.djoining parts of the drive is temporarily lost when the transmitted force is rcversed; as a consequence the surfaces formerly exerting opposite forces ;\II~ llloving apart, until contact is again made at the other side of the gear l.ooLIt. Clearly this discontinuity of the transmitted force, possibly combined wi I.h Coulomb friction, is in conflict with a smooth and precise position con I.ml [1365]. Backlash can be avoided by employing high quality mechanical ("()I II jlOllents, but wear cannot be excluded. A mechanical solution would be 1111' IIse of spring-Ioaded gears with split wheels which maintain a bilateral (·olll.ad force irrespective of the sign of the transmitted load force. AIlother solution requiring only standard gears is to split the drive into : ;('V(~ral Illlits and to preload the gears electrically; this is of particular interest ii' til(' type of load makes a distribution of the driving force desirable anyway. Ali (~xample could be the azimuth motion of a large satellite antenna, i.e. the j'()l.al.ioll around the vertical axis; antennae of this type require very accurate jlosil.iouing, for example with a final tolerance of less than 1/100°. This rorn:spollds to a resolution of the fulI circle to one part in 36,000 which ddillitely calls for direct measurement by a digital sensor mounted at the pivoL !>caring of the main construction. The two or more servo motors could I II : arranged at symmetrical points arollnd the circumference and exert torqlle I.hmlll·;1t pillioll drives to the central gear. A simplified schematic is drawlI ill I"i/':. I ri. 1:\ ass1lmiu!; two motors, with the main strueture of thc étutellUi1 (tlld 1.11(' I.wo drivI~ lIIoLors I>eillg IUlI1ped illto thr<'(~ sqmrate rigid I>o
15.2 Linear Position Control
381
the low torque region, where backlash would be most troublesome. As the required load torque mM Ref increases, the motor that initially opposes the motion, eventually reverses its torque and comes to the help of the other motor until there is nearly equal load sharing at full torque. Centralload mass
m c t, ro l,fl
mC2'
~
Motor 1
ro2' f2
Motor 2
Fig. 15.13. Position drive with two servo motors and electrically preloaded gears for counteracting the effects of backlash
When assuming symmetry between the two motors, neglecting ali fric tional torques and referring all angles, angular velocities, inertias and torques to the central axis, the block diagram in the right hand side of Fig. 15.14 de scribes the mechanics of the drive. The torque control loops are again replaced by equivalent lags having unity gain. ln addition to inherent damping effects not shown in Fig. 15.14, such as speed dependent frictional torques, active damping will definitely be needed with this complicated mechanical structure. Two kinds of oscillations are likely to require countermeasures Gears wi/h elasticity and backlash
Fig. Hí.'14. Silllplificd block diagram of three-mass twin drive with preloaded gears for CO\lllt(·l'Ilc'l.ill/', hacldlLHh
15. Some Applications of ControJled Electrical Drives
:\X:2
15.3 Linear Position Control with Moving Reference Point
383
Y~ '~
• oscillations of the two drive motors in phase opposition while the central mass remains at rest, • oscillations of the two motors in phase against the central mass.
M
1
mM
E
Provided the torque control loops respond sufficiently fast, both oscilla I.iolls can be damped by adding suitable signals to the torque references, LlmMl Ref
= -
LlmM2ReC=
DI (Wl - W2) - D 2 (Wl
+ W2
-
D I (WI-W2)-D 2 (Wl+W2-
2w) , 2w).
T!te signals for the motor speeds w}, W2 can be derived from sensors attached 1." the motor shafts. It is noted that the steady state characteristics of the dri ve system are not altered by these auxiliary feedback loops because the ~il-!;Ilals vanish in steady state under symmetrical conditions [T31]. Fig. 15.15 demostrates another pair of nonlinear functions that would be ati vantageous for the preloading of gears because they cause the motors to produce equal torque at fullload while maintaining constant overall gain in I.!t(! speed control loop. Again these nonlinear functions would be cumber ~O lllC to realise with analogue circuitry but they present no problems when Illicrocomputers are employed for the controI.
/
/---
a
x
Contjnuou~ I Coordinate : rBferenc~s m: transformation I cartBSian I coordinates :
JRof
~
:
Motion in polar coordinates
Decoupled conlrol in polar coordinatBs
!
:' l;;+;t--~,~r ,·~t--~~""'I + lx<+i' -
~
b r ro2 8C:C~::~~ff~
I
!t:ir x<+i' -ll--=-r,~"---
, Motion in f cartesian : coordinatBS
,, I I I
,, r
I
arctan{
b
Fig. 15.16. Dual axis position control in polar coordinates with reference signals supplied in cartesian coordinates (a) Mechanical modelj (b) Block diagram
/-mM tRel + mM 2 ReI
mM ReI /
/ /
"' il~.
tó.15. Torque functions for electrically preloading of of gear train
t\llother example of a dual axis position controlled drive is shown in Fig. I;). 1fi a. It could be part of a machine tool or a robot with a mechanical :,t rllcLul'c in cylindrical coordinates r, f, z, where each axis is activated by a :;('par;\I:e position controlled drive having an inner torque controlloop with !.II(! eqllivalent lag Te. This is a simplification of the example discussed in (:hap. 2. For the particular application it could be necessary to generate I.!t, ! r<'f(,fCllce trajectory in x, y, z-coordinates, while only machine orientated 'I',' , ;;-1I1Ci\S Urements are available. Clearly the motion in the z-direction is de , '0 11 pkd ir it is vertically orientated so that gravitational effects are exclndcd I 1>11 t ii, trallsfornmtion bctwcen (7', f) and (x, y) is nceded. It can be proclll'ecl I,,Y I. r
drive coordinates. The set (x, y, Z )Rej would again be produced by a suitable reference generator. Dynamic coupling effects between the r, f motions caused by centrifugaI and Coriolis-forces can be approximately compensated by adding suitable corrective terms to the torque references, as was mentioned beforej the ef fectiveness of this method of compensation, called inverse load modelling, depends of course on a fast response of the inner controlloops. With the present state of microelectronics alI the control functions seen in Fig. 15.16 b can easily be executed with 1 kHz sampling frequency, which would be adequate in most cases.
15.3 Linear Position ControI with Moving Reference Point ln Sect. 15,1 it was mentioned that there are cases, for example with paper mills, where the speed of each drive section must be controlled with high precis ion in relative synchronism to the speed of a neighbouring section. This task cnu
384
15.3 Linear Position Control with Moving Reference Point
15. Some Applications of Controlled Electrical Drives
lt is based on a conventional analogue speed control loop using, for ex ample, a De tachometer as a speed sensor. Due to unavoidable drift of the analogue components there is a normal control error, usually below 10- 2 of Ilominal speed, which could, of course, be reduced by one or two orders of Illagnitude, but only with considerable effort, such as thermal stabilisation of the criticaI components, and at high cost. The idea of the scheme in Fig. 15.17 is to add a digital corrective channel for compensating the steady state error of the analogue speed control loop; rapid response is not important since a well-tuned analogue speed control leaves little to be desired in this respect.
ZI
385
Zl
Overflow
z(v+1)
Fig. 15.18. Programmable frequency divider
ln order to generate a finely adjustable reference frequency h having the sarne steady state accuracy as the fixed clock frequency io derived from a Quartz oscillator, the programmable frequency divider shown in Fig. 15.18 might be used. It contains an arithmetic unit for executing a recursive addi tion according to Digital
controller
Inkremental sensor
Fig . 15.17. Quasi-continuous digital speed control
The digital controller compares the pulse train h from an incremental where each pulse corresponds to an angular increment .dE, with a Il'krence pulse train fI derived from a high accuracy frequency source. A I( ~ v( ~ rsible counter registers the forward and reverse pulses; if none is lost, the t'()I\Lcnt of the counter is a measure of the angular error of the drive shaft wiLh respect to the reference angle. With the help of a digital-to-analogue COIl verter it is transformed into a small corrective signal .dWl for the analogue Sl)(~ed loop. Hence the digital channel may be understood as an integral speed conLroller without drift or as a proportional position controller [J24, L31] .
z(v
+ 1) =
[z(v)
+ zd mod
Zmax,
(15.6)
where Zmax is the capacity of the register while Zl < zmax corresponds to the desired reference frequency fI established by the overflow pulses. The addition is synchronised with the clock frequency io. The mean balance of the continued addition is
S('IISor ,
fI
zmax
=
io Zl
so that a pulse train results, the mean frequency of which is adjustable in fine steps and has the sarne precision as the constant clock frequency io, Zl
fI= -io·
(15.7)
Zmax
The overflow pulses are not quite equidistant but when operating the fre quency divider at a sufficiently high clock frequency, this inherent phase jitter has no effect on the speed of the motor [L32]. Similar methods are used in the process industry for in-line blending, where the flow rates of liquids to be mixed in prescribed ratios are measured by tnrbine wheel flow meters whose speed is transmitted and processed in the fmm of plll s (~ trains [R12]. A 1\ adjllsl.;\,hi<- n'fcr<~llCC frc(jll(:llcy lJlay abn llC formed with the help of ;~ plta:w IlII.I'
:186
15. Some Applications of Controlled Electrical Drives
15.3 Linear Position Control with Moving Reference Point
multiple of the constant clock frequency, f = zlfo, which may then be rcduced by a scaler with capacity Zmax' ln various production lines where continuous strips or webs ofmaterial are hcing processed in subsequent stations it is often specified that the stations .~hould be decoupled for preventing tension in the material. For this purpose a limited storage of material is accumulated in the form of a loop between t.he stations which, by taking up the difference of incoming and outgoing IIIaterial, represents a measure of the synchronism between the two adjoining drives. Naturally, one should attempt to maintain the loop at approximately constant mean length to have adequate reserves in both directions, should t.he speeds of the two drives exhibit temporary fluctuations. This is illustrated in Fig. 15.19 a showing the principIe of a magnetic tape drive as used on computers. The incoming tape is fed from a reeI and pressed a.gainst the drive rolls AI, A~ to be transported with high acceleration and velocity past the read-write head S; subsequently the tape is rewound on a nother reeI driven by a separate motor. By keeping the tape between the reels taut, the maximum acceleration would be severely limited by the inertia of the reelsj also, excessive force in the tape could not be excluded. ln order 1.0 avoid these problems, loops are formed on both sides of the drive rolls which are kept staight with moderate forces by mechanical or pneumatic equipmentj this greatly reduces the effective inertia and the applied forces, when the tape is accelerated. The more slowly accelerating reeI drives must of course produce sufficient maximum speed for preventing the storage loops from reaching their upper or lower limito
ii
387
Only the left half of the tape drive in Fig. 15.19 a is considered in the following, with the tape velocity VI at the read-write head S assumed to be an independent (impressed) variable. The speed V2 of the feed rolls should be so controlled as to maintain the length X2 of the storage loop at a prescribed vaI ue. Because of dX2 =
dt
t (V2 - vd
(15.8)
the block diagram in Fig. 15.19 b results, where the storage represents an integrator with two inputs cancelling in steady state. Normally the feeding reeI is controlled to maintain constant tension of the unwinding tape. The quality of the control can be further improved by introducing to the speed controller of the feed rolls A 2 a feed-forward signal from the velocity VI, which acts as a disturbance. ln view of the quickly changing operating conditions it could even be appropriate to provide acceleration feed-forward to reduce the necessary length of the storage loop. It is noted that the in ertia of the reeI drive depends on whether the spool is full or nearly emptyj also, most magnetic tape drives are reversible, i.e. they can operate in either direction.
I
Ifm .1.
b
b
Fig. 15.20. Geometry of loop in a continuous rolling mil!
a Feed - forward
\
I
.----..,
-I~I
./
X2
v2
I
ali l : . . i : ----
VI
. 1
mL
mI\1R~A:11
a2
On multi-stand continuous rolling mills the storage loops between suc cessive stands, which are necessary to trim the coordinated speed control described in Sect. 15.1, have a different geometry, as is shown in Fig. 15.20; an electrically or hydraulically powered looper roll pushes the material with a force h upwards and thereby keeps the strip taut. Due to the geometry of the loop there is a nonlinearity between the deflection x and the length of the loop , I = .Jb2 + x 2 leading to a nonlinear differential equation 2b::.d(xjb) Vo I dt
/) Jo'ir~ . 1 r;. I {lo I'·h:d d ri v" fo r maglldi c, t. np c
whcr<:
VII
V2_VI Vo Vo
(15.9)
coulel oe thc rated velocity; the integrating time constant
(1\) Schcllml,icj (J.) Blnel, di ugrn.m T,
'J. ii
:1:
'/I"
I
(I~U())
15. Some Applications or Controlled Electrical Drives
:\88
15.4 Time-optimal Position Control with Fixed Reference Point
dcpends on the ratio x/I, i.e. the shape of the loop. x/L = 1 corresponds to the special case in Fig. 15.19 a. Operating the loop at smaU values of x / l becomes increasingly difficult because the integrating time constant is reduced; this calls for fast response of the looper and the adjoining main mill drive used for controlling the loop. The force 1m exerted on the material is aIso a function of the loop geom ctry; in relation to the vertical force h exerted by the looper roll, it is
1m IL
1 I
v
km/h 120
;dso indicating a criticaI range at small values of x/I, where the gain becomes very large. A similar problem as with loop control arises with automated tracked vchicles developed in recent years for urban transportation, Fig. 15.21 a. The distance x of each vehicle to the leading vehicle must be controUed to a speed dependent safe value XRef by applying driving or braking force 1M ~ O. The distance can be measured, for example, by radar or by sensing equipment ill stalled in the guideway.
. . _--_ ......
60
....
041------,-------,-------__
(15.11)
= 2 x '
389
o x
10
20
tis
= xRei =SOm '........\>,
""
O-t O
"'-"""'"
"
--...-"
f MRef
20
tis
Fig. 15.22. Braking transients of distance control scheme --~-
- - - - - - ------,,-
fM
'L X
a fL
VI
A computed transient of the distance control scheme is shown in Fig. 15.22; it may be caused by a sudden constant deceleration of the leading vehicle; a PID-controller and a linear function XRef(V2) were assumed for the example. Transients of this sort and their dangers are well known to the motorist; the task of maintaining a safe distance can be alleviated by employing advance signals, such as feed-forward from the deceleration of the leading vehicle or the rate of distance change, as is common practice in road traffic by use of braking lights or warning signs.
15.4 Time··optimal Position ControI with Fixed Reference Point b
Fil-:. 15.21. Distance control of tracked vehicles
(1\) Sct.cmatic; (b) Block diagram
W IWII again
n~ g arding
the vclocity of the forward vchicle as an indepen tlw control systcm for the distallce between t he vchidcs ("( )illd h av I: t.l1(' fOl"I1\ ShOWll ia Fig. 15.21 b. li: J"('prc:;en(.s a cascadcd s dl( ~Jll.t: " urIÜ tillilll': il \l l<'L" 1
t ll l l ' ·. ·· .1'1 1,., 1,0 U H' " '/".d iul'. vdliclc i n 1':('n•.' l'Il. L~'d mi " r
l\
d, ·l ny,'d III Ild i 11(';1.1.' J'lIllcLÍ.OII
V<'!f1 ('1I. V, 111 1 " KI II II. I"I, · II I'l l r,ll ! H,J ' 1 1\.j,l l' l'o vi lJl lI J.l II I \ I l lli1 r .. " r! t.ld ll l'. di :il. l l.lIl ·l '
Some applications of drives call for the discontinuous motion of a machine part from one steady state position to another. Examples are boring machines where the tool must be positioned for the next drilling operation or reversing rolling mills where the upper rolI is lowered by a screw-down drive as soon as the slab of m aterial has left the roU gap. Since the load is disengaged during the traveI, thcre is no specified trajectory x(t) ; the only requirement is that tIl<: driv(~ sllOlIld follow tlw discontiJlllOIlS position reference in minimum time l: iv ('" I.he po we r ra.t.ill l'; nr til(' driv(' alld w il.hollL ovr rshooting the targcl: pO: ií l.illll ,
15.4 Time-optimal Position Control with Fixed Reference Point
15. Some Applications of Controlled Electrical Drives
:\90
With the assumption that
391
fL
• the response time of the torque control loop can be neglected, • the jerk, i.e. the derivative of acceleration, need not be restricted (which would be necessary if a transportation system with on-board passengers or a weakly damped mechanical structure were involved), • the frictional torques are ignored, I.he drive with the associated mechanicallinkages can be approximated by a double integrator having the acceleration a(t) as input and the position x(t) itS output, as seen in Fig. 15.9. The well known time-optimal solution that could be derived with the cal culus of variations, consists of two intervals of maximum accelerationjdeceler ittion and possibly an intermediate interval with constant maximum velo city. This is shown in Fig. 15.23 for two different step changes of the reference po sition [38,44]. Velo city and position are plotted versus time as well as in state space which, because x and vare the only state variables, reduces to a state pl ane. Beginning at t = O in steady state condition, v(O) = x(O) = O, a new position reference is commanded, XRef = XOl, calling for a new target state 'II (t e ) = O, x (te) = XOl with te --+ mino The time optimal transient consists of lII aximum initial acceleration, operation at maximum speed and controlled deceleration to the prescribed target point; the maximum values of acceler ation and deceleration are assumed equal in Fig. 15.23. If the amplitude of t,he position step is reduced, X02 < XOl, the maximum velocity may not be r('ached; the acceleration is then immediately followed by deceleration with I.he switch- over taking place at half the distance travelled.
v- Ti<
a amax
-amax
t3 X0 1
x
v
x X OI
F Il{. 1r....~ :L Id ' ·ld in,·,1 Lim.,'-o p (I,iJlIl..! 1"'~II, ionill,{ (n) ~ H, J t..h · 1' 1t111t' 1 (J., ) 1t1')l t ) ( IH I U ~ vrf. 1,llIln
!.n l ll!l;" U (,
wil.h lilll ikd 1lc<~c:I" r1~l.i o ll
Fig. 15.24. Time-optimal position control scheme
ln the state plane, the intervals of constant acceleration or deceleration are characterised by parabolae marked with time, the interval at constant velocity appears as a straight line parallel to the x-axis. The multi-Ioop position control scheme in Fig. 15.24 closely models the idealised behaviour shown in Fig. 15.23 [A24] . The nonlinear position con troller may be realised digitally to achieve reproducible accuracy at high resolution, for example 0.1 mm with a maximum distance of 1 m on a screw down positioning drive. The scheme differs from the linear one in Fig. 15.9 by the nonlinear function in the position controller, VSoll
= v 2 a m ax lei
sign e,
(15.12)
which generates the velo city reference as a function of the remaining position error e; its purpose is to guide the drive with constant maximum decelera tion into the reference point. The mode of operation, after a step change of the position reference has occurred, is then as follows: When XRef changes by avalue exceeding twice the maximum braking distance, the velo city ref erence VRe f initially assumes its maximum value V max causing the velocity controller to become saturated and the drive to start at maximum acceler ation. When the velo city reaches the reference value, the velocity controller reverts to linear operation and maintains maximum velo city. As the position error is gradually reduced, VRef(e) follows the parabolic function, resulting in constant deceleration until the reference position is reached. The drive may then be disconnected by disabling the power supply and arrested with a mechanical brake. This is necessary because of instability due to the infinite gain of the position controller at e ~ O. ln case the position reference changes by less than twice the maximum braking distance, the drive will switch from acceleration to deceleration before the maximum velo city is reached. The torque control loop in Fig. 15.24 serves only for protection in case of overload; the rnaximum current should be adequate to produce maximum acee! craLion a t fnU load torqlle. 1\ lIo(.\wr 1I1 () f( ~ <'! e f.Çi.tIlt llldhod for n' a.lising lninirnum time position con l.rol ili ·,d" 'w II i II II'i , ~ . I fi, 2;' , wh( ~ J'(~ i.l\( ~ lill<'ilí' r d c r f'll('(' Kiglln.l w ~ncrnL()r of I,'il',' I ~, . H 1, 1111 1" ,(, 11 1I""lili"d l." ,',"1\1'1':.1.1.,' 1.1", r, ',·,I-r"rwnr, 1 ;; i,.,lI i t.l: l )'I"llIi!" ',1
15. Some Applications of Controlled Electrical Drives
:\!)2
I -. X 7 I~arget X
V Ref
Flef
15.4 Time-optimal Position Control with Fixed Reference Point
DC or AC drives having an effective lag of the torque loop of a few ms. A time-optimal transient of an AC servo drive was shown in Fig. 14.9; some simulated transients , assuming different initial conditions, are plotted in Fig. 15.26.
a Ref
v(x)
Speed restríction vmax at x > x 1
Fig. 15.25. Dynamic reference modei for minimal-time position control Braking curve
time-optimal transients. The advantage of this arrangement is that the coutrol system proper remains unchanged, i.e. linear and optimally tuned. 'I°lte choice, whether the transients should be linear or time-optimal is solely ddermined by the reference model; this structure is particularly suited for illlplementation on a microcomputer. As a further variation, the nonlinear function in Fig. 15.25 is modified around zero (in dashed lines) to prevent overshoot and to achieve a stable nmtrol mode at the reference position. The advantages of both control strate l\i es are then combined: For large position error, the drive performs nearly I.ime-optimally while it becomes linear once the position error has entered a Ilarrow range around steady state condition; the sarne could be applied to Lhe position controller in Fig. 15.24. It has been mentioned that the second order approximation is only appli cable as long as the residuallag of the torque controlloop can be neglected. ' I'hus t.he application of this control scheme is restricted to converter supplied
II
ror
Fttrr
~Rel
!
t
I I
I
~ I I
t
~ I I
I
I I
I I
~
-r;-
X
Slnle plano
IVX I .II" I
I I
~Posítion
v~ v~
F ig. H •. :.W. M i n l l.II".! 1'.i l lL'·
x
t
tt.
~
I'0 ~ il.i,,"i" g 1.1'Il1I::i'·IIt.1I wi t.h difr"lt ll d. ill i t.il\l (·olldil.ioIlH
393
------, "
I I
e
I I
,
I I
'
I
"
:
~!-v x=o
,
x
1
I
\ \
I
'\ 1"x2 , ,~~;:x2
x
Position of traín
at begín of brakíng
Fig. 15.27. Controlled deceleration of automatic train along a veocity profile
Controlled deceleration pro files are also used on other drives such as ele vators or automatic trains, where at any time a fictitious braking curve v(x, t) must be known, along which the vehic1e could be decelerated or brought to standstill with the maximum specified deceleration. By continuously compar ing this projected velo city profile with possible forward speed restrictions, it can be determined, when braking should begin to meet the constraint. Fig. 15.27 illustrates the situation, where an automatic train with its projected velo city profile approaches a restricted speed zone. Whenever passengers on board vehic1es are involved, it is usually speci fied that the accelerating forces must change gradually to avoid discomfort or anxiety (as every passenger standing in a bus knows)j a similar situation exists, when complex mechanical structures such as satellite aerials or robots are to be moved, because abruptly changing acceleration or deceleration cre ates unnecessary forces and oscillations. This calls for smooth but swift posi tioning, still using all the available drive power. Some idealised time-optimal positioning transients with limited rate-of-change of acceleration (jerk) are drawn in Fig. 15.28, beginning with large change of position reference (a) which is gradually reduced to case (e); (b) and (d) are limiting cases sepa rating different modes of operation. Clearly, the condition that da/dt should be finit.c~ muounts to the definition of acceleration a(t) as another continuous st a tp va.riahk j IWllce, the state v(~ctor [a(t), v(t) , x(t)] could be represented as a cOIILillll()II:4 I.r;l,j"" .I.ory ill 1t 1'. 111"('( ' dillH'Jlsiolla,1 Spi\.CP .
394
15.4 Time-optimal Position Control with Fixed Reference Point
15. Some Applications of Controlled Electrical Drives
t5
", V Ref
"
395
VRef
Vmax
", t8 = O t aRef
= - a max
a U""~"''<'"''''
/'
uí"''''''>
,::uwt~uaRef"-.YRef "' " ,
t7
b
XRef - XTarget
Fig. 15.29. Switching curves and trajectories in state-plane
c
VRef
X Ref
VRef
aRef
t
e
d
aRef
F ig. 15.28. Idealised time-optimal positioning transients with limited jerk
Fig. 15.30. Dynamic modeI for minimum time, limited jerk, positioning transients
As this is diflicult to visualise, some characteristic curves are plotted in I"ig. 15.29 in the x, v-plane, showing how the controlled deceleration phase 01' the transient could be implemented; the starting phase is no particular problem because it follows the simple strategy of applying maximum da/dt ll11lil the peak values of acceleration and, possibly, velo city have been reached. Figure 15.30 shows yet another modification of the dynamic model to 11(: used for generating the reference functions [a(t), v(t), X(t)]Ref for time opLilllal positioning. The acceleration reference is now the output of another iIlL(:grator, making it a continuous state variable of the dynamic model. The rd'(:I'('I1CC variables are again to be used as inputs for an optimally tuned liJlellr Ululti-loop position control scheme as seen in Fig. 15.9. The nonlinear
is so chosen as to generate in each interval a target acceleration aT which the acceleration reference then approaches with constant slope. Clearly, the main advantage of control with the help of a dynamic refer ence model is that the computation of the complex set of reference functions can be performed with an idealised mathematical model that is not affected by the unavoidable approximations when tuning the real drive control system. As is indicated in Fig. 15.30 it is, of course, necessary to bring the initial conditions of the mo deI in line with the plantj this would best be done in standstill, i.e. with x(O), v(O) = a(O) = O.
rllllcLioll
(/('(i)
f(I', 'U, a)
(I:í.J :1)
15.5 Time-optimaI Position ControI with Moving Reference Point
15. Some Applications of Controlled Electrical Drives
:\%
15.5 Time-optimal Position Control wit h Moving Reference Point III some applications the task of positioning in minimum time is made more ('olllplicated by the fact that the reference point itself is moving. This could aSS1lme the form of a rendez-vous-problem, where the drive is supposed not olll,V to reach the reference point in minimum time but to assume also the S ;llll C final velocity as the target. The minimum time requirement could be dllc t.o the condition that the initial state of the drive should be subject to I"ast restrictions for reaching the rendez-vous position in time. Examples for this type of problem are the control of a robot with the Laslc of gent1y picking up a fragile object on a moving conveyer or of a fl.ying slt CiH' that is to cut a strip of metal, emerging from a rolling mill at fulI speed, into various custom-Iengths sections. Similar problems exist also with rotary printing presses, when a new supply rolI of paper must be attached to Lhe end of the outgoing rolI while the press is running; the situation vaguely ['( ~sembles a relay race, where the new runner must start in time for accepting Lhe token from his predecessor in a specified space, or driving on the ramp to :1. 1II0torway, where the entering vehicle is merging into a moving line of cars.
dynamic performance of the drive; with DC machines this condition was sometimes met with twin motors for reducing the effective inertia. There is no limiting specification on the jerk in this application, except the finite ris e time of the motor torque.
a Accelerating
Braking
(previous run)
Motion of target
~1' J
V
--- . ,r I
i
a
'.
+
Corrective interval
\
x
.,x
I
/
. / .......
X2
Next ILIrgot
V
"" .-.J...
l. 1
(!2Gr-2
t = t2
X, (t)
t= t
•.
>'
/" 1
lnitial position
of shear
çt)V2
I c:
x 10
~I
I"ig. 15.31. Principie of a fiying shear
'l'he basic layout of a rotating shear is sketched in Fig. 15.31. The strip- or rod-shapccl material coming from the last stand of a rolling mill moves with vdoci l,V /I j through the shear, which consists of two mechanicalIy coupled drlllw; lH!aring opposing cutting blades. InitialIy the shear may be at rest; :d. a pl'(~ddcnllined time tI is is accelerated, 50 that the tools separate the lII : d.(~rial al a. prescribcd point. The blades and the material should move in :l.ppr(\ X illlal.( ~ s,VI\ChrOllislll during the scparation, lmt a slightly higher vdoc.. iLy 01' 1.\1(' I.ools cunl.! also bc sp ecified for separatillf, t.he ma.terial. F()lI()willl~ UI<' ('1011. , 1.1 1.(' l'i lwitt' iR decderabxlmu.1 lIsu;dly sd a(. Ii(~sl. aI. a S(·.J('c/:cd w;ül.. j l'l ~ p,,, J l. ioll . II. i ~ r llcl.inV; a lld !' ~ ';)h q.r "'( ,d I II I.('IUI I,h n ll I.w.. 1.'<,·\I ,,JIIt. HI\l 1I "I' 1111' /l 111' l u \... llidl n dh 1'''1' 11I1 ', h
397
b } I
X1
/
/
/
Fig. 15.32. Discontinuous motion of fiying shear
Beside this discontinuous mode of operation there is also a continuous mode, when, due to lack of time, the shear moves at a pulsating speed without coming to standstill; the reason may be increased velocity of the material or the need for shorter cuts of material. An idealised accelerating, cutting and subsequent positioning cycle is shown in Fig. 15.32 with the variables plotted vs. time and in the state plane; the circular motion of the shear is represented by the circumferential path of the blades. Time and position are defined relative to the cut, i.e, the acceleration phase takes place in the negative range of time and position; the material velocity is assumed to remain constant after the start of the shear, VI = consto The shear begins to accelerate with a = a max at t = tI; at t2 it reaches cutting speed and then continues in synchronism with the material until the cut occurs at t = O. The interval t 2 < t < O serves for correction of velo city and position to eliminate errors caused by tolerances and the finite response time of the speed loop. Because of the assumed constant velocity of the material the position Xl (t) of tltc refcrcnce point changes linearly with time, while the shear position :1:2(1) ro]J()W S iII I.l\(~ intcrval tI < t < t 2 a parabolic function. During the ;1.1'
cc -11'I':t1. i() II P li ii.: iC'
15. Some Applications of Controlled Electrical Drives
398
to.
= t2 -
tt
a
15.5 Time-optimal Position Control with Moving Reference Point
V1
(15.14)
max
the reference point moves by Xl
(t 2 )
-
Xl (tl)
=
VI
to. =
V2
(15.15)
_1_ .
a max
The shear velocity rises linearly with time in this interval
V2(t) = (t - td amax =
t - tI t2 - tI
VI - - - ,
tl:S t
:s t2 ,
(15.16)
399
The control of the shear drive is facilitated by the use of a suitable dynamic model in combination with a high performance multiloop position control scheme as in Fig. 15.9. The dynamic model in Fig. 15.33 is of the type shown in Fig. 15.25 producing the reference signals for minimal time positioning transients; the only difference is a feed-forward signal of the target velocity VI, for example coming from measuring rolls. There are two modes of operation, intermittent and continuous motion. aReI
Acceleration
which results in a necessary distance for acceleration 1
2
1
v;
(15.17)
X2(t2) - x2(td = "2 a max (h - td = -2 - - ; a max
t.his is half the distance covered by the target point during the sarne intervalo The remaining corrective interval is travelled in approximate synchronism,
Xdt2)
(15.18)
= X2(t 2) = VI t2 < O.
While it would also be possible to determine the instant when the shear should start at maximum acceleration by a time-based criterion, it seems more appropriate to use a closed loop control scheme for velo city and position which remains in operation during the complete cycle until the shear comes to rest again; this has the advantage that tolerances such as a slightly varying I.arget velo city can be compensated during the final part of the approach 11316, K69, L39, 07]. X2(f l )
continuous operation intermittent operation
Fig. 15.34. Simulated transients of time-optimal shear drive
= f(v l )
• ln reset of the intermittent mo de the feed-forward signal is removed and the position reference corresponds to the starting position X20 which is a function of target velo city
Measuring rolls vI' MI ~vI
-V=-
2
X20
= - ~+V1t2. 2a max
(15.19)
This condition (Reset and Wait) is maintained until time tI when the target position counter reaches the starting position count Xl(tl)
X;. " n(
V;.
/I,.,
,"li"
nM
Firl ' Ilr••:I :1. II YII '"l1 k 111",1<-1 1'111' (" "ll. l'ulli''' H f~ 1I .1' 1I1'H 11 1",,,1'
V2
= __1_ a max
+ VI t2.
(15.20)
• At. t.his time tI the velocity feed-forward signal is added. As a consequence, t.h(~ driv(' st.a.rts with maximllm acceleration. The position reference input ror 1.1... III(ld,·1 i:; III1W dcriv( ~ d rrolll lhe tarl';et position counter which is ('()IIIII.,·.] l.el lV lil d ll '/.( ' 1'11 h-y 11 11/,. 11 ln" i!wJ('IIi( ·III.:; (Ir 1.11(' t/IC'WillJ'('llll'lIl:; l'olls.
15 . Some Applications of Controlled Electrical Drives
100
15.5 Time-optimal Position Control with Moving Reference Point
• As soon as the target position counter reaches zero and the cut occurs, the mo deI is again put into reset while the desired length XlO of the next section is entered into the target position counter as an initial condition for the next run. (:lw.nging the operating mode is indicated in Fig. 15.33 by mechanical switches; in reality this would of course be done electronically or by soft \vare.
It can be shown that the mo deI in Fig. 15.33 produces time-optimal posi
transients as long as the physical restrictions imposed by the drive are exceeded. Some simulated transients are depicted in Fig. 15.34 where the ()p(~wting conditions are changing due to a reduction of the desired section lellgth; this results in a transition from discontinuous to continuous mode. ' I 'he transients prove that the control scheme can in principIe meet these dif liclllt specrncations; it is of course important to select a drive that is able to ntpidly follow the reference functions . I.iolüng
1I0t.
81
,
.. . . ... ':;..' Speed [100min .-.1] .... / . ....
6
I ' I' . I·I
'I
.
....
\'
\.
.•.. .. ..\:,i~~.\......;, ~~ ....
..//...'r '
/ /
(I
I
\
I . ~ ~'\
\
\, .
.
'" .. ...\
I
/ :
forque [kNm]
.
','
1
I
.
\.
I' : ' . . " Ir :'"
, I
1
'I
':
\, _ / :
''\
:
I I
.
.
....:'>-- .... - - - - - -
"\" . ,1:... ... \"'" \. \:" .'j;;":_
I':
I' I ..I,'
'
\.:
I
/.
:\ I' \ I . '\ ." " . . \ ... . )~.......... . ... . '\
Motor current [kAJ
:~.,.
'- -.I"I'
.. .. . . -:. . • .. •. . .. :.. . ... ·1·· .:- .... . . ...:- . .. .. ... .;. . . . . . .
·1" " .
. (j
ii
I
o
0.1
i
i
0.2
0.3
i
0.4
0.5
0.6
0.7
sec 0.8
Fig. Ui.3Ó. Mcas ured iransients of a flying shear drive with a 500 kW induction '-illpplicd by voltage source GTO- inverter and field orientated control .
/IIoto\"
1':vI 'lI ~HI illl.I ~ l"lllil.l . (~llt drives of this killtl, calljllf~ for lal'(~c overload capaI> i l.y 11.11" f'/l,Il l.t'fll. IH Iil:ü hlc r :~H pOIlH~~ , coul.rolkd. A<; nlol.o('s ;Ll'< ~ 1I0W tll l<.:<':CSSflllly
il lll'I'Ü'Il !.i IlI'.
PC 11 1111. .. 1'11; l.I\('i l' n'''! w~loI i llt'rfI.ill lu 1111, 11.01<1"" lI.dvalll.il.r~(' . Ali (' X II, Ftl'. I'! ,.:\f. nlt"whlf( 1·(... ·" ld,·oI 1,1'1 111 111" 111. )1 (Ir 11 ll,Y hlJ': 11 111' /11' d .. ivl'
" I Ir~tI., · b llil1 ' \\I
401
during an acceleration and repositioning cyc1e. The 500 kWinduction motor is supplied from a 800 kVA GTO volt age source inverter and has a control structure as shown in Fig. 12.28 [K75J. The accuracy ofthe materiallengths to be expected from this type of drive control depends of Course on many factors such as accuracy and response of the drive controlloops, accuracy and resolution of the measurements from the moving material as well as the length of time t 2 available for correction during the final approach. Typically, some cm tolerance could be expected, when cutting material of 10 m length. The task of controlling a flying shear is quite complex; on the other hand, the gain in productivity is substantial, when comparing it with a shear that can cut only when the material is at resto The next step towards fully automatic operation is on-line minimisation of scrap by computing an optimal combination of different custom lengths sections that can be produced from a given size ingot. AlI these control func tions , the flying shear control with a dynamic reference model as well as the optimisation can be executed today on a single board standard microcom puter.
Bibliography
Books and Proceedings
[19]
Amin, B., Induction motors, Analysis and torque control. 2nd Ed., Document Service, Paris, 2000 Bedford, B.D., Hoft, R.G., Principies of inverter circuits. Wiley, New York 1964 Bird, B.M., King, K.G., An introduction to power electronics. Wiley, New York 1983 Bõdefeld, Th., Sequenz, H., Elektrische Maschinen. 8th Edition, Springer, Berlin-Heidelberg-New York 1971 Bose, B.K., Power electronics and AC drives, Prentice Hall, Englewood Cliffs, N.J . 1986 Bühler, H., Einfiihrung in die Theorie geregelter Gleichstromantriebe. Birkhiiuser, Basel-Stuttgart 1962 Bühler, H ., Einfiihrung in die Theorie geregelter Drehstromantriebe, Bd. 1 Grundlagen, Bd. 2 Anwendungen. Birkhiiuser, Basel-Stuttgart 1977 Bühler, H ., Regelkreise mit Begrenzungen. VDI Fortschrittsberichte, Reihe 828, Bd.8, VDI Verlag Düsseldorf 2000 Buxbaum, A., Schierau, K ., Berechnung von Regelkreisen der Antriebstech nik. Elitera, Berlin 1974 Concordia, C., Synchronous machines - Theory and performance. Wiley, New York 1951 Davis, R.M., Power diode and thyristor circuits. Peter Peregrinus, 1971 De, G., Electrical drives and their controls. Acad. Books Ltd, Bombay-New Delhi 1970 De Almeida, A., Bertoldi, P., Leonhard, W. (Eds.), Energy efficiency im provements in electric motors and drives, Springer Verlag Berlin Heidelberg New York 1997 Dewan, S.B., Straughen, A., Power semiconductor circuits. Wiley, New York 1975 Eckhardt, H., Grundzüge elektrischer Maschinen. Teubner, Stuttgart 1981 Ernst, D., Strõle, D., Industrieelektronik. Springer, Berlin 1973 Finney, D., The power thyristor and its applications. Mc Graw Hill, New York 1980 F'rõhr, F., Orttenburger, F., Technische Regelstreckenglieder bei Gleich strom-Antrieben. Siemens AG, Berlin-München 1971 GroB , n., Slllle, G., Elektrisdle VOrf;chlllmllt.rieb.~ fiir Werkzeugrnasclliuen.
[20]
SicmcllH 1\(:, Jkrlill~Miin d l"ll 10~1 CYllli,yi, II>., 1','lIny, B.I1., S l.al.ic I"'w"!' 1'1'<''III''II''Y ('''all l\ , ~ r !l. WiI. ~y , Nnw York
[1]
[2] [3]
[4] [5] [6] [7] [8]
[9] [10]
[11] [12] (13] [14] [15] [16] [17] [18]
11I71i
illl~
Bibliography
Bibliography
Habinger, E., Die Technik der elektrischen Antriebe. VEB Technik, Berlin 1979 Inl Hamming, R.W., Numerical methods for scientists and engineers. Mc Graw Hill, New York 1962 I·.ul I-Ieumann, K ., Grundlagen der Leistungselektronik, 2. Aufl. Teubner, Stutt gart 1978 1·/tl l I-I eumann, K., Stumpe, A.C., Thyristoren. Teubner, Stuttgart 1969 1.. "1 I-Iofmann, A., Stocker, K., Thyristor-Handbuch, 4. Aufl. Siemens AG, Berlin München 1976 1·!.Ii! I-10ft, R.G., Semiconductor power electronics. Van Nostrand Reinhold 1986 I~~71 Jõtten, R., Leistungselektronik I. Vieweg, Braunschweig 1977 I.. . KI Jordan, H., Weis, M., Asynchronmaschinen, Vieweg, Braunschweig 1969 1'2!)1 Kessler, C. (Ed.), Digitale Signalverarbeitung in der Regelungstechnik. VDE-Buchreihe Bd. 8, VDE, Berlin 1962 1:\0] Kleinrath, H., Stromrichtergespeiste Drehfeldmaschinen. Springer, Wien 1980
1:\ LI Kovacz, K.P., Racz, J ., Transiente Vorgãnge in Wechselstrommaschinen.
Ung. Akad. d. Wiss., Budapest 1959 U ., Stephan, R.M., Microcomputer-based adaptive control applied 1:t21 Keuchel, to thyristor-driven DC-motors. Springer, Berlin-Heidelberg-New York 1993 I:U I I<ümmel, F ., Elektrische Antriebstechnik. Springer, Berlin-Heidelberg-New York 1971
Laible, Th., Die Theorie der Synchrorunaschine im nichtstationãren Betrieb.
I:\~ I Springer, Berlin 1952 1:\:,1 Langhoff, J., Raatz, E., Geregelte Gleichstromantriebe in der Praxis der Stromrichter- und Regelungstechnik. Elitera, Berlin 1977 1:1(i! Lappe, R., Thyristorstromrichter für Antriebsregelungen. VEB-Technik, [ler! in 1970
1:\71 J,p.onhard, A., Elektrische Antriebe, 2. Aufl. Enke, Stuttgart 1959
I:IKI L<mlhard, W., Einfiihrung in die Regelungstechnik, 6. Aufl. Vieweg, Braunschwcig 1991
I:I!II Lcollhard, W., Regelung in der elektrischen Antriebstechnik. Teubner,
S(,\lI.l.gart 1974
1'1111 1.
II "AC Symp., Düsseldorf 1974, VDIjVDE, 1974
I.collhard, W. (Ed.), Control in power electronics and electrical drives. 2''''
I'"I II''AC Symp., Düsseldorf 1977, Pergamon Press, Oxford 1977 Lconhard, W. (Ed.), Microelectronics in power electronics and electric:tl 1/121 drives. ETG-Fachberichte Vol. 11, VDE, Berlin 1982 Lconhard, W. Digitale Signalverarbeitung in der Mess- und Regelungslcch 1'1:1I Ilik. 2. Aufi. Teubner, Stuttgart, 1990 Lemer, A.J., Schnelligkeitsoptimale Regelungen. (Translation from RII ~ ' I'I~ I sian), Oldenbourg, München 1963 I'I!,I Lyon, W.V., Transient analysis of alternating-current machinery. MIT PrcHll alld J. Wiley & Sons, New York 1954 I"(i! Mad,i, K., Winograd, O., Stromrichler. Oldcnbourg, München 19:J:3 lVIa:ld<~, F.F., Thyristor control. Wilcy, New York 19n i'l'71 IVI"YI'r, M., Selbslgeführte Thyrislor-Stromridllcr, 3. AIlIl. Si(:III"">I AI:, I'IKI IkrliIl,. Miinchcn 1971 I'WI M'·,II.I;(:\I, (:., Nd~p;efii!trlc Slrolllricht,l:r /llil, 'l'hyrisLort:/I, :1. Anil. Si, '11, <:., :'I.,o,"ri.-lIi,('I'l,(·('IIl1 il" 1': i,"'il1t1I1I1I', iII Wirllllll lV'1 WI'I H(' ,,".1 'l'h"1l ~ I,( II
114' H i(IIIH'II': !\<: I t('rlill I q,q:~
I~ Ij
I
[51]
405
Mohan, N., Undeland,T.M., Robbins, W .P., Power electronics, Converters, applications and design, J. Wiley, New York 1989 [52] Mohr, O. (Ed.), Steuerungen und Regelungen elektrischer Antriebe. VDE Buchreihe Bd. 4, VDE, Berlin 1959 [53] Murphy, J.M.D., Thyristor control of AC motors. Pergamon Press, Oxford 1973 [54] Murphy, J.M.D., Turnbull, F.G., Power electronic control of AC motors. Pergamon Press, Oxford 1988 [55] Novotny, D .W. , Lipo, T.A., Vector control and dynamics of AC drives. Clarendon Press 1996 [56) Orttenburger, F., Das regelungstechnische Verhalten der Gleichstrommas chine . Siemens AG , Berlin-München 1972 [57] Pearman, R.A., Power electronics: Solid state motor control. Prentice Hall, 1980 [58] Pelly, B.R., Thyristor phase-controlled converters and cycloconverters. Wi ley, New York 1971 [59] Pfaff, G., Regelung elektrischer Antriebe. Oldenbourg, München 1971 [60] Pfaff, G., Meier, Ch., Regelung elektrischer Antriebe II, (Geregelte Gleich stromantriebe). Oldenbourg, München 1982 [61] Puchstein, A.F., Lloyd, T.C., Conrad, A.G., Alternating current machines. Wiley, New York 1954 [62] Rajashekara, K., Kawamura, A., Matsuse, K ., Sensorless control of AC motor drives . IEEE Press, 1996 [63) Say, M G., Alternating current machines, 5th Ed. Pitman, London 1983 [64] Say, M.G., Taylor, E.O., Direct current machines, Halsted Press 1980 [65) Schõnfeld, R., Digitale Regelung elektrischer Antriebe. Hüthig, Heidelberg 1990 [66] Schrõder, D., Elektrische Antriebe, Bd. 1-4, Springer Verlag 1995 [67] Seefried, E., Müller, G., Frequenzgesteuerte Drehstrom- Asynchronantriebe. Verlag Technik, München 1992 [68] Sen, P.C., Thyristor DC-drives. Wiley, New York 1981 [69] Sen, P.C., Principies of electrical machines and power electronics. Wiley, New York 1989 [70] Simon, W., Die numerische Steuerung von Werkzeugmaschinen. Hanser, München 1971 [71] Slemon, G.R., Straughen, A., Electric machines. Addison Wesley, 1980 [72] Spãth, H., Steuerverfahren fiir Drehstrommaschinen. Springer, Berlin-Hei delberg-New York-Tokyo 1983 [73] Steimel, K., Jõtten, R., (Hrsg.), Energieelektronik und geregelte elektrische Antriebe. VDE-Buchreihe B 11, VDE, Berlin 1966 [74] Steven, R.E., Electrical machines and power electronics. Van Nostrand Rein hold, 1983 [75] Taegen, P., Einfiihrung in die Theorie der Elektrischen Maschinen I, II. Vieweg, Braunschweig 1971 [7()] Tschilikin, M.G., Elektromotorische Antriebe. (Translation from Russian), VEB-Technik,1957 I77J lJhlmann, E., Power transmission by Direct Current. Springer, Berlin Ilcid elberg-New York 1975 I,na,] Vas, 1'., Veclor control of A C m(Lcll i1\ ("s. Clarendon Press, 1990 17K "1 Va,;, I'., SI:II~orlp.ss v(:c:I,M illlel di!'I'I '1. \'
Bibliography 'HJG
Bibliography
VEM-Handbuch - Die Technik der elektrischen Antriebe. VEB-Technik, Berlin 1979
Vogel, J., Elektrische Antriebstechnik. 5th Ed. Hüthig, Heidelberg, 1991
IXIJ Th., Schaltungslehre der Stromrichtertechnik. Springer, Berlin Wasserrab, IX'2J
Gottingen-Heidelberg 1962
Watzinger, H. , Stromrichter-Gleichstromantriebe. Hüthig, Heidelberg 1980
IR:\] Weber, W ., Adaptive Regelsysteme . Oldenbourg, München 1971
I X ~J IX:;J Weh, H., Elektrische Netzwerke und Maschinen in Matrizen-Darstellung. Teubner, Stuttgart 1968
Wood, P., Switching power converters, Van Nostrand Reinhold 1981
IX(>] IX7J Yamamura, S. , AC motors for high performance applications, analysis and control. Marcel Dekker, New York 1986
Zadeh, L.A., Desoer, C.A., Linear system theory. McGraw Hill, New York
IXKj 1963
IXO]
Papers, Surveys and Dissertations Abbondanti, A., Method of flux control in induction motors driven by vari able frequency, variable volt age supplies. Proc. Int. Semiconductor Power
Converter Conf. Rec. 1977, pg. 177
IA 2] Abbondanti, A., Brennen, M.B., Variable speed induction motor drives use
electronic slip calculation based on motor volt ages and currents. IEEE Trans. Ind . Appl. 1975, pg. 483
a squirrel-cage motor fed
IA :1] Abraham, L., Transient characteristics and limits of nd
from a frequency-converter with DC current link. 2 IFAC Symp., Control
in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 475
Ii\ 'II Abraham, L., Heumann, K ., Koppelmann, F., Wechselrichter zur Drehzahl
sl.cncrung von Kafiglaufermotoren. AEG Mitteilungen 1964, pg. 89
I i\ ~,I Abraham, L., Koppelmann, F., Kafiglaufermotoren mit hoher Drehzahldy
lI illllik. AEG Mitteilungen 1965, pg. 118
I/I(i I AIon' ns, D., Günther, H., Elektrische und mechanische Einflu13gro13en bei
I, a.~ercgelkreisen. AEG Mitteilungen 1976, pg. 249
tu :arnley, P.P., French , C.D ., Torque estimation and control in PM motor
1 1\ '/ I
drives. Proc. EPE 95, Sevilla, pg . 3.411-3.416 IMI Ackva, A., Binder, A. Greubel, K. Piepenbreier, B., Electric vehicle drive with surface-mounted magnets for wide field weakening range. Proc. EPE 97, Trondheim, pg. 1.548-1.553 IA')1 Ahrens, D., Raatz, E ., Regelungstechnische Untersuchungen von Antrieben mil. Kupplungs-oder Getriebelose. AEG Mitteilungen 1973, pg. 210
IA I() I AilllC, M.L., Degner, M.W., Lorenz, R.D., The effects of saturation induced
sali cllcy movement on flux angle estimation. Proc. Workshop on Advanced
MoLion Control (AMC) 98, Coimbra, pg. 369-374
IAI 1] Aillsworth , J .D ., Harmonics instability between controlled static convertors
il11(1 AC-n etworks. Proc. IEE 1967, pg. 949
IA I '21 Alcagi, II ., Nabae, A., A new control scheme for compensating the torquc
I. rallsfcr fllncLion of a self-controlled synchronous motor. IEEE Trans. Ind. API'I. HHH, P!;, 795
IA 1:1] Alhn:cht , P., Dic gcn:gdl,e oOPl'dt!;es(l( ·isl\', AHynd lfllllmascltille alH drdl'~
'"rdtlvill'm,I,I,',r (:"II('fal.o r fHlI Ndz. DillH , 'I'll 11\'1tI11I ~:chw( : i K 1!)R1
[AiJ
407
[A14J Albrecht, P., VolIstedt, W., Microcomputer control of a variable speed
doubly-fed induction generator operating on the fixed frequency grid. Mi
croelectronics in Power Electronics and Electrical Drives, Darmstadt 1982,
ETG-Fachberichte 11, pg. 363
[A15) Alexanderson, E .F.W., Mittag, A.H., The thyratron motor. AIEE Trans.
1934, pg. 1517
[A16] Aller Castro, J.M ., Montilla, A.B., On-line parameter estimation of the in duction machine model using active and reactive power balance for spatial vectors of field oriented drives. Proc. EPE 97, Trondheim, pg. 4.609-4.614 [A17) Ambrozic, V., Klaassen, H., A modified method for the determination of the
rotor time constant of an induction motor in a cold state. Proc. 8 th EDPE,
Pula, 1994, pg. 66
[A18) Amler , G., Ein neues Steuer-und Regelverfahren für einen Stromzwischen
kreis-Umrichter mit abschaltbaren Leistungshalbleitern. Diss. TU Braun
schweig , 1991
[A19] Amler, G., A new control method for current source inverters with self
extinction devices, combining low distortion of voltage and current with fast
dynamic response. Proc. EPE 91 , Firenze, pg. 3-211
[A20) Andresen, E. C., Pfeiffer, R ., Werth, L., Fundamentais for the design of high
speed induction motor drives with transistor inverter supply. Proc. EPE 89,
Aachen, pg. 823
[A21) Andresen, E. C., Gediga, S., Schwartz, H. J., A continuous PWM-square
wave transition method for voltage vector control of induction motors. Proc.
EPE 91 , Firenze, pg. 2-156
[A22) Andresen, E. C., Haun, A., Influence of the pulse width modulation control method on the performance of frequency inverter induction motor drives. Europ. Trans . on Electr . Power Engng. 1993, pg.151 [A23) Andria, G., Salvatore, L., Signal processing for determining flux and torque
of induction motors. Proc. IEE-Power Electronics and Variable Speed Drives,
London 1988, pg. 296
[A24) Anke, K., Ertel , K., Sinn, G., Digitale Wegregelung. Siemens Zeitschrift 1960,
pg. 664
[A25) Anke, K., Kessler, G ., Müller, H., Digitale Drehzahlregelung. Siemens
Zeitschrift 1960, pg. 660
[A26J Antic, D., Klaassens, J.B., Deleroi, W., An integrated boost-buck and matrix
converter topology for low speed drives. Proc. EPE 93, Brighton, pg. 5-21
[A27) Appt , D ., Baisch, R ., Bromer, G., Wegmessung und Wegregelung an nu
merisch gesteuerten Werkzeugmaschinen. Siemens-Zeitschrift 1974, Beiheft
Steuerung und Antriebe zur Automatisierung der Werkzeugmaschine, pg. 12
[A28) Appun, P ., Control of rotating and linear induction motors for vehicle
drives . 2n d IFAC Symp., Control in Power Electronics and Electrical Drives,
Düsseldorf 1977, pg. 809
[A29) Armstromg, G .J ., Atkinson, D.J., A comparison of model reference adaptive
system and extended Kalman filter estimators for sensorless vector drives.
Prec. EPE 97, Trondheim, pg. 1.424-1.429
[A30) Arnet, B ., Jufer, M., Torque control on electric vehicles with separate wheel
drives . Proc. EPE 97, Trondheim, pg. 4.659-4.664
[A31) Arsudis, D ., Doppeltgespeister Drehstromgenerator mit Spannungszwischen krci~-Umrjchter im Rotorkreis für Windkraftanlagen. Diss. TU Braun ~d lw(' i (,l;
f9R9
[A: 12 1 Ar:1I 1.dí li , I). , Vollstedt, W., Sr:mlOrl ()S ( ~ Rcgclung einer doppeltgespeisten A' :Ylll'Ior""I11Il II,'Ioilll: mil. g( : rílll~"ll Nd.:r.riickwírkungen . Archiv für Elek f. r,d,, ','llll ll(, 10U I ) pg. X~)
~08
Bibliography
l!d3] Atkinson, D.J ., Acarnley, P.P., Finch, J.W., Parameter identification tech niques for induction motor drives. Proc. EPE 89, Aachen, pg. 307 1/\34] Atkinson, D .J., Acarnley, P.P., Finch, J.W., Method for the estimation of rotor resistance in induction motors. Proc. EPE 91, Firenze, pg. 3-338 1/\ 35] Atkinson, D.J., Acarnley, P.P., Finch, J.W., Observers for induction motor state and parameter estimation. IEEE Trans. Ind. Appl. 1991, pg. 1119 IIIJ6] Attaianese, C., Damiano, A., Marongiu, L, Perfetto, A., A DSP-based non linear controller for induction motor drives. Proc. EPE 95, Sevilla, pg. 3.914 3.918 1/\:17] Attaianese, C., Damiano, A., Marongiu, 1., Perfetto, A., Identification of rotor parameters and speed estimation in induction motor drives. Proc. EPE 97, Trondheim, pg. 4.615-4.619 1/\:\8] Attaianese, C., Tomasso, G., Perfetto, A., Position control of induction mo tor by means of vectorial torque controI (VTC) strategy. Proc. EPE 99, Lausanne 1111] Bachmann, G., EngeI, D., GeregeIte GIeichstrom-Hauptantriebe bis 400 kW für Werkzeugmaschinen. Siemens Energietechnik 1981, pg. 269 1112] Bader, W., Theorie der SchaItungen für die Widerstandsbremse von selbster regten Gleichstrom-Reihenschlu13motoren bei elektrischen Fahrzeugen. Wiss. Verõff. Siemens Konzern 1930, pg. 209 lln] Balestrino, A., De Maria, G ., Sciavicco, L., Adaptive controI design in ser vosystems. 3rd IFAC Symp. ControI in Power Electronics and Electrical Drives, Lausanne 1983, pg. 125 \lH] Bannack, A., Nguyen Hong Ha, Riefenstahl, D., Speed and angular synchro nism controI of machines with flexibIy coupled drives. Proc. EPE 95, Sevilla, pg. 3.464-3.468 11\:'1 Barrenscheen, J ., Flieller, D., Kalinowski, D., Louis, J.P., A new sensorless sp eed and torque controI for permanent magnet synchronous motors: Real isation and modelling. Proc . EPE 95, Sevilla, pg. 3.839-3.844 11\(i] Ilassi, E ., Benzi, F., Scattolini, R ., Design of a digital adaptive controller for d ectrical drives in industrial appIications. IEEE Trans. Ind. Electr. 1992, p~. 120 I\;í.~si, E., Benzi, F.P., Bolognani, S., Buja, G.S., A field orientation scheme 1 H7 1 for current-fed induction motor drives based on the torque angle c1osed-Ioop ("oIlLro!. IEEE Trans. Ind. Appl. 1992, pg.1038 IHKI lIaucr, F., Heinle, G., Eliminating the harmonics from measured current values in PWM drives. Proc. EPE 91, Firenze, pg. 3-343 \lWI Ibyer, K.H ., Waldmann, H., WeibelzahI, M. Die TRANSVEKTOR-Re gdung für den feldorientierten Betrieb einer Synchronrnaschine. Siemens Zeitschrift 1971, pg. 765 III I 01 Bayer, K.-H. , Blaschke, F., Stability problems with the controI of induction utachines using the method of fieId orientation. 2nd IFAC Symp., ControI in I'ower Electronics and Electrical Drives, Düsseldorf 1977, pg. 483 III J II llecker , H., Dynamisch hochwertige DrehzahIregelung einer umrichterge Hpcisl,l:JI Asynchronmaschine. Regelungstechnische Praxis 1973, pg. 217 II \ 121 Ikckl'r, n., ROBOT CONTROL, eine Mikroprozessorsteuerung fúr Indus I.ri "roboLer. Sicmens Energietechnik 1980, pg. 137 IIII:\[ Ik illldI' 1,;1'1'; ~JI , 11;1"1 '11,," , PJ',. :\:W
Bibliography
409
[B15] Belmans, R.J.M., Verdyck, D., Geysen, W., Findlay, R.D., Electromechan ical analysis of the audible noise of an inverter-fed squirreI-cage induction motor. IEEE Trans. Ind. Appl. 1991, pg. 539 [B16J Bender, K., Pandit, M., Weber, W., Verlustoptimale Regelung von rotieren den Scheren. Regelungstechnik 1970, pg. 540; 1971,S. 8 [B17] Bender, K ., BeschIeunigungsgeregelte Antriebe in optimalen RegeIkreisen. Regelungstechnik 1972, pg. 423 [B18] Ben Ammar, F . Pietrzak-David, M., de Fornel, B., Mirzaian, A., Field oriented control of high-power induction motor drives by Kalman filter ob servation. Proc. EPE 91, Firenze, pg . 2-182 [B19] Ben Ammar, F . Pietrzak-David, M., de ForneI, B., Mirzaian, A., Power range extension of an induction motor speed drive by using a three-IeveI GTO inverter with space vector modulation. Proc. EPE 93, Brighton, pg. 5-219 [B20] Ben Brahim, L., Kawamura, A., Digital current reguIation of field-oriented controlled induction motor based on predictive flux observer. IEEE Trans. Ind. Appl. 1991, pg. 956 [B21] Ben Brahim, L., Kawamura, A., A fully digitised field-oriented controlled induction motor drive using only current sensors. IEEE Trans. Ind. Electr. 1992, pg. 241 [B22] Ben-Brahim, L., Improvement of the stability of the V If controlled induction motor drive system. Proc. IECON 98, Aachen, pg. 859-864 [B23] Ben-Brahim, L., Tadakuma, S., PracticaI considerations for sensorless in duction motor drive system. Proc . IECON 98, Aachen, pg. 1002-1007 [B24) Bergin, T. J ., Cholewka, J., The direct Ioad angle control of a CSI drive. Proc. IEE-Power EIectronics and Variable Speed Drives, London 1988, pg. 305 [B25] Bernot, F ., Gonthier, L., Bocus, S.D., Elbaroudi, S., Berthon, A., Kauff mann, J.M., High efficiency drive for electric vehic1es permanent magnet motor fed by a two-stage converter. Proc. EPE 97, Trondheim, pg. 1.554 1.559 [B26] Bianchi, N., BoIognani, S., Parasiliti, F., Villani, M., Prediction of overload and flux-weakening performance of an IPM motor drive: analyticaI versus finite element approach. Proc. EPE 99, Lausanne [B27] Bichler, D., Synchronmaschinen mit aktiver Dampfung. Diss. TD Braun schweig 1985 [B28) Bilewski, M., Fratta, A., Giordano, L., Vagati, A., Villata, F., Control of high-performance interior permanent magnet synchronous drives. IEEE Trans. Ind. Appl. 1993, pg. 328 [B29] Binns, K.J ., Wong, T.M., Analysis and performance of a high-field perma nent-magnet synchronous machine. IEE Proceedings 1984, pg. 252-258 [B30) Blaabjerg, F. , Pedersen, J.K., Digital implemented random moduIation strategies for AC and switched reluctance drives. Proc. IECON 93, Maui, pg. 676 [B31] Blaschke, F ., Ripperger, H., Steinkõnig, H., Regelung umrichtergespeister Asynchronmaschinen mit eingepragtem Standerstrom. Siemens Zeitschrift 1968, pg. 773 [B32] Blaschke, F., Hütter, G., Scheider, D., Zwischenkreisumrichter zur Speisung von /\synchronmaschinen für Motor-llnel Generatorbetrieb. ETZ-A, 1968, 1'1 (· IOK
[I\:B] Illr\w lll«',
I,'., Ila:i I'rin7.ip der 1", · 'dlll·it'lll.i ~ I'III1~, die Cntndlajie riir díc 'I 'H li. NSV " ;1';','( )I!: II "f:d II III', VIIIl /\" ,I' III"II"""III;I.", 'lIill"lI . Si':/ll( :/l'; Z. :ik:chrin, 11)'/ I"'I ', 7f,';
~IO
Bibliography
Bibliography
111:\4] Blaschke, F., The principie of field orientation as applied to the new
TRANSVEKTOR closed loop control system for rotating field machines.
Siemens Review 1972, pg. 217
111:lS] Blaschke, F., Das Verfahren der Feldorientierung zur Regelung der Asyn
chronmaschine. Siemens Forschungs-und Entwicklungsberichte 1972, pg . 184
II \:\G] Blaschke, F ., Das Verfahren der Feldorientierung zur Regelung der Drehfeld
maschine. Diss. TU Braunschweig 1973
11\:17] Blaschke, F., Bõhm, K., Verfahren der Fluflerfassung bei der Regelung strom
richtergespeister Asynchronmaschinen. 18t IFAC-Symp., Control in Power
Electronics and Electrical Drives, Düsseldorf 1974, Vol. 1, pg. 635
I' \:\K] Blaschke, F., Strõle, D., Transformationen zur Entflechtung elektrischer
Antriebsregelstrecken. Regelungstechnik 1973, pg. 105
!lU!)] Blasco-Giminez, R., Asher, G.M., Sumner, M., Rotor time constant identi fication in sensorless vector control drives using rotor slot harmonics. Proc. EPE 95, Sevilla, pg. 1.083-1.088 111'10) Bodson, M., Chiasson, J., Novotnak, R., A systematic approach to selecting
fiux references for torque maximization in induction motors. IEEE Trans.
Control Syst. Techn. 1995, pg. 388
111~ I] Bodson, M., Chiasson, J ., Novotnak, R ., Nonlinear speed observer for high
performance induction motor control. IEEE Trans. Ind. Electr. 1995, pg.
337
11I~ 2 ] Bocker, J., JaIUling, J., Discrete-time flux observer for PWM inverter fed
induction motors. Proc. EPE 91, Firenze, pg. 2-171
II I~:l] Bohm, E., Führung und Regelung von Antennenantrieben fiir Satelliten
Bodenstationen. Siemens Zeitschrift 1974, pg. 828
I'I,I~] Bõhm, K., Wesselak, F., Drehzahlregelbare Drehstromantriebe mit Um
, richterspeisung. Siemens Zeitschrift 1971, pg. 753
IH'I:,) Boehringer, A., Knõll, H., Transistorschalter im Bereich hoher Leistungen
und Frequenzen. ETZ 1979, pg. 664
Ill'H>] Iloehringer, A., Stute, G ., Ruppmann, C., Vogt, G., Würslin, R., Entwick
IllTlg eines drehzahlgesteuerten Asynchronmaschinenantriebs für Werkzeug
In ; ~'ie hinen . Zeitschrift für industrielle Fertigung 1979, pg. 463
I' 1,1'( I Hiillillg, B., Leonhard, W., Propulsion control of a track-powered linear syn ('lirollo us motor for a magnetically levitated vehicle on the basis of power 1I1(:;.l:;urements in the inverter station. Proc. Int. Conf. on Electrical Ma d,ill(~S, Brussel 1978, paper L 5/1 III'IHI I luers ting, H., Knudsen, M., Vadstrup, P., Standstill estimation of electrical p
11I(~c:lt a lliq\le d'une machine asynchrone d'induction. Conumel 83, Toulouse
PI~ . IV-43
Im, ii Iloroj"vi c, D., ltobust nonlinear control algorithm for fast positioning in
>:<'I.'vo drivcs. Proe. EPE 89, Aachen, pg. 1375
!I li,'.! I n o> :<" LUC, Adjl.lstable speed AC drives -- a teehnology status r eview. Prol:.
11 ': 1':1': I !lH'l" P(~. II()
I'11, :q 110:1" , n.IC(I':c1.), MicroC()lTIpllt( ~ r ('oul,rol nr 1)()W' ~ r d('d rouicH alld driv eR .
11':1':1':, N,',w YIlI'It IHH'7
[111,'1[ lI ullC' , 11.1( ., 1'''w' 'I> ,·I(·c l.ro" i ... ; ILllcI IIl,d.i '>l 1 " .. ,,101,,1 T( ·(·ltllolol\'y 'II' , , . " " 1,, " ,,01 ;1 11,:1':1': '1'''" 111 I,," i\ 1'1'1 1\l0: 1, 1'1'. . IIII ',!
nl.lI.I.II: : lU ,d
411
[B55) Bottura, C.P., Augusto Filho, S.A., Robust torque tracking control for the
induction machine. Proc. IECON 98, Aachen, pg. 1533-1537
[B56] Bowes, S.R., Bird, B.M., Novel approach to the anaIysis and synthesis of
modulation processes in power converters. Proc. IEE 1975, pg. 507
[B57] Bowes, S.R., Midoun, A., Microprocessor impIementation of new optimal
PWM switching strategies. Proc. IEE 1988, Pt. B. pg. 269
[B58) Bowes, S.R., New regular-sampled harmonic elimination PWM techniques
for drives and static power converters. Proc. EPE 91, Firenze, pg. 3-241
[B59] Bowes, S.R., Advanced regular-sampled PWM control techniques for drives
and static power converters. Proc. IECON 93, Maui, pg. 662
[B60) Bowler, P., Carter, R.C ., Jones, D.M., The development and performance of
a digital subharmonic modulator for AC drives. Proc. IEE-Power Electronics
and Variable Speed Drives, London 1988, pg. 403
[B61) Braga, G., Farini, A., Fuga, F., Manigrasso, R., Synchronous drive for mo
torised wheels without gearbox for light rails system and electric cars. Proc.
EPE 91, Firenze, pg. 4-78
[B62] Brandes, J ., Greubel, K., Winter, U., Antriebe für Elektrofahrzeuge - ein
Konzeptvergleich. etz 1994, H. 15, pg. 846-851
[B63) Brandenburg, G., Trõndle, H.-P., Das dynamische Verhalten des Regis
terfehlers bei Rotationsdruckmaschinen. Siemens Forschungs-und Entwick
lungsberichte 1976, pg. 17-20 und pg. 65-71
[B64] Brandenburg, G., Wolfermann, W., State observers for multi-motor-drives in processing machines with continuously moving webs. ProC. EPE 85, Brussels, pg. 3.203-3.210 [B65] Brandenburg, G ., Schãfer, U ., Compensation of Coulomb friction in in
dustrial elastic two-mass-system through adaptive control. Proc. EPE 89,
Aachen, pg. 1409-1415
[B66J Brandenburg, G., Papiernik, W., Feedforward and feedback strategies ap
plying the principie of input balancing for minimal tracking errors in CNC
machine tools. Proc. Workshop on Advanced Motion Control (AMC) 96,
Tsu City, pg. 612-618
[B67) Brandenburg, G., Geiflenberger, S., Kink, C., Schall, N-H., Schramm, M.,
Multi-motor line shafts for rotary printing presses - a revolution in printing
machine techniques. ProC. Workshop on Advanced Motion Control (AMC)
98, Coimbra, pg. 457-462
[B68] Brandstetter, P., Cermak, T., Chlebis, P., Kozina, J., Stefanik, M., Micro
computer system control in field coordinates of synchronous machine. Proc.
EPE 91, Firenze, pg. 4-560
[B69] Brandstetter, P., Mech, M., ControI methods for permanent magnet syn
chronous motor drives with high dynamic performance. Proc. EPE 95,
Sevilla, pg. 3.805-3.810
[B70) Bressani, M ., Odorico, A., Sica, M., High-speed, large-power induction mo
tors for direct coupling to variable-speed gas compressors. Proc. EPE 93,
Brighton, pg. 5-366
[B71) Brickwedde, A., Head, D., Graham, H ., Microprocessor controlled 50 kVA
PWM inverter drive. IEEE Ind. Appl. Ann. Meet. 1981, pg. 666
[B72] Brickwedde, A ., Microprocessor-based adaptive control for electrical drives.
3,· 01 IFAC Syrnp. Control in Power Electronics and Electrical Drives, Lau
Slllllll' 19H3, pg. 119
11l7:1j. Ilr'il'l(weddl', A., Sdb~t e iIl H l.dkJld(·r, oll--lill" adal'tiver R egler auf Mikrorech 'lI'rI, lt" j: , l'ii r .-I1 ·Id.I·i:;dw AIlt,ri, ·I.... Ili:.: i. '1'\1 II rrwlIschw uig 19!1 !) I1I'l-1 I 1I ";;llrt'·, 1\ ., "'·IIII, ·I,,'rl'.'·'· , ( L , 1\1 (' 1'111' '', '1''' . , l' I>lIl l.i ,, " ill \ Itt'll rn/''y 0(' 1L NI'II Hll r 1<-111 1 ""II.,,,II, ·oI 'II" ' vI> oI rl v,' ".\' 11 1,,'," I' , ,,, 1':1'1': Ui> , S.. vi ll " , 1'1'. ' :\ . 107 :1,1 7'1.
,112
Bibliography
Bibliography
II ln] Brosse, A., Henneberger, G., Sensorless control of a switched reluctance mo t or using a Kalman filter. Proc. EPE 97, Trondheim, pg. 4.561-4.566 II \76] Brosse, A., Henneberger, G., Schniedermayer, M., Lorenz, R.D., Magel, N., Sensorless control of a SRM at low speeds and standstill based on signal power evaluation . Proc. IECON 98, Aachen, pg. 1538-1543 jl m ) Brown, G.M ., Szabados, B., Belmans, R.J.M., High power cycloconverter drive for double- fed induction motors. Proc. EPE 91, Firenze, pg. 2- 282 11\7~) Brunotte, C., Schumacher, W., Detection of the starting rotor angle of a PMSM at standstilL Proc. EPE 97, Trondheim, pg. 1.250-1.253 III? t)) Brunotte, C., Regelung und Identifizierung von Linearmotoren für Werk zeugmaschinen. Diss. TU Braunschweig, 2000 II \~()) Bühler, E., Eine zeitoptimale Thyristor-Stromregelung unter Einsatz eines Mikroprozessors. Regelungstechnik 1978, pg. 37 \ II~ I) Bühler, H., Investigation of a rectifier regulating circuit as a sampled data system. 4 th IFAC-Congress, Warsaw 1969 jI 1~2) Bühler, H., Umrichtergespeiste Antriebe mit Asynchronmaschinen . Neue Technik 1974, pg. 121 \ Il~: l) Bünte, A., Grotstollen, H., Parameter identification of an inverter-fed induc tion motor at standstill with a correlation method. Proc . EPE 93, Brighton, pg. 5-97 \I IK;j) n ünte, A., Grotstollen, H., Off-line parameter identification of inverter-fed
induction motor at standstilL Proc. EPE 95, Sevilla, pg. 3.493-3.496
IllH!i) Buja, G.S., Indri, G.B., Optimal pulse width modulation for feeding AC
motors. IEEE Trans. Ind. AppL 1977, pg. 38 \I\~( i\ Buja, G.S., Fiorini, P., Microcomputer control of PWM inverters. IEEE 'I'rans. Ind . Electr. 1982, pg. 212 I, :;:,7 1 Iluj a, G.S. , Menis, R., Valla, M.!., Disturbance torque estimation in a sen sorlcss DC drive. Proc. IECON 93, Maui, pg. 977 i' \H HI Ilussc , A., Holtz, J., A digital space vector modulator for the control of ;1 Lllrcc-ph ase power converter. Microelectronics in Power Electronics and 1':I ,,,: l.ri ca.l Drives, Darmstadt 1982 ETG-Fachberichte 11, pg. 189 li \ii, <) I 1111 xl >i111l!l, A ., Regelung von Stromrichterantrieben bei lückendem und lIiclli.liickelldem Ankerstrom. AEG Mitteilungen 1969, pg. 348 111 ' 1I1j IItl x1"~IIIlI , A. Aufbau und Funktionsweise des adaptiven Ankerstromreglers. i\ 1':<: M ilteilungen 1971, pg. 371 111111\ 1\ II X haum, A., Einsatz von adaptiven Reglern bei geregelten Stromrichter Sl.t· llgliedern. VDE-Aussprachetag Freiburg 1973, pg. 99 III: 1','1 I\ysLr un, K., Strom-und Spannungsverhaltnisse beim Drehstrom-Drehstrom lllllri chler mit Gleichspannungs-Zwischenkreis. ETZ-A, 1966, pg. 264 111\):1\ lly sLron, K., Meissen, W., Drehzahlsteuerung von Drehstrommotoren über /,wischenkreisumrichter. Siemens Zeitschrift 1965, pg. 254 \<: II ( :;lIubronne , J.P., Semail, B., Rombaut, C., Vector control of a PWM current stlllrCi> inverter-fed induction motor. Proc. EPE 91, Firenze, pg. 2-177 I( :z\ (:;tpolin o, G.A., Du, B., Extended Kalman observer for induction machine rolor currents. Proc. EPE 91, Firenze, pg. 3-672 1< ::1\ CardolcUi, L., Jufer, M., Sensorless position detection using the supply volt agl: for ii. programmable current drive for synchronous motors. Proc. EPE !lI, J"ir(~ fl z c, pg. 4-J 23 ICljl (:ard tl.';o, F.D.S ., Marlins, J.F' ., Pires, v.r., A comparative study of a PI, 1(("11 mi 1Il'l.work 1\11.1 Fuzzy [~l~nclit approncb e()Jll. rnll<~rs for a AC driv( ~ . Proc. W"r!(:: I\f'p
011
IIdvlIllct :d Mot.ioll COIILrtll (II M<:) !lK, Ctlilllbra, PI~ .
:r7:"
:I~ O
[C5) [C6] [C7] [C8) [C9)
[CIO)
[Cll) [C12) [C13) [C14) [C15) [C16) [C17) [C18) [C19) [C20) [C21)
[C22) [C23) [C24)
413
Castillo-Castaiieda, E., Okazaki, Y., Load variation compensation in real time for motion accuracy of machine tools. Proc . Workshop on Advanced Motion Control (AMC) 98, Coimbra, pg. 305-309 Case, M.J., Kulentic, P.K., A microprocessor controller for the cyclocon verter. Microelectronics in Power Electronics and Electrical Drives, Darm stadt 1982, ETG-Fachberichte 11, pg. 171 Case, M. J., Van Wyk, Transputer based induction motor fiux-control method. Proc. EPE 91, Firenze, pg. 2-192 Cataliotti, A.,Tecniche di controllo dei motore asincrono - Un nuovo schema di controllo vettoriale encoderless. Diss. Univ . Palermo 1998 Cecati, C., Rotondale, N., On-line identification of electrical parameters of the induction motor using RLS estimation. Proc. IECON 98, Aachen, pg. 2263-2268 Chapuis, Y.A ., Girerd, C., 'Aubépart, F ., Blondé, J.P., Braun, F., Quantiza tion problem analysis on ASIC- based direct torque control of an induction motor . Proc. IECON 98, Aachen, pg. 1527-1532 Chauprade, R, Abbondanti, A., Variable speed drives: Modem concepts and approaches. Proc. Int. Semiconductor Power Conv. ConE. 1982 , Ind. AppL, pg. 20 Chen, H.H., Microprocessor control of a three-pulse cycloconverter. Proc. IECI 1977, pg. 226 Chiasson, J. , Dynamic feedback linearization of the induction motor. IEEE Trans. Aut. Contr. 1993, pg. 1588 Chiasson, J ., Nonlinear controllers for induction motors. IEEE Trans. Aut. Contr. 1995 Chiba, A., Fukao, T., A closed loop operation of super high speed reluctance motor for quick torque response. IEEE Trans. Ind. Electr. 1992, pg. 600 Choi, J .S., Kim, S.U., Lee, E.J., Kim, Y.S " A sensorless vector control of the saturated induction motor using third harmonic voltages. Proc. IECON 98, Aachen, pg. 977-980 Chollet, G., Commande par microprocesseur d'un commutateur de courant pour moteur asynchrone de traction. Conumel 83, Toulouse pg. IV-1 Claussen, U., Adaptive zeitoptimale Lageregelung eines linearen Stellantrie bes mit synchronem Linearmotor. Diss. TU Braunschweig 1979 Claussen, D., Fromme, G., Motorregelung mit Mikrorechner. Regelungs technische Praxis 1978, pg. 355 Claussen, D., Leonhard, W., Microprocessor-controlled linear synchronous motor as positioning drive. Proc. Int . ConE. on Electrical Machines, Brussel 1978, L 5/2 Clénet, S., Vinassa, J.M., Astier, S., Lefevre, y" Lajoie-Mazenc, M., Infiu ence of the brushless DC motor torque compensation by acting on the current waveshapes on the torque speed characteristics. Proc. EPE 95, Sevilla, pg . 3.898-3.902 Colak, L, Garvey, S., Wright, M.T., Mixed-frequency testing of induction machines using inverters. Proc. EPE 93, Brighton, pg. 5-317 Collings, T.D., Wilson, W.J., A fast response current controller for micro processor based SCR-DC motor drives. IEEE Trans. Ind . AppL 1991, pg. 921 Conrath, C., Commande á. microprocesseur d 'une machine asynchrone au L()pil o Ux~. Application à un sysLerne de levage. Conumel 83, Toulouse, pg. I V ,7
-11-1.
Bibliography
Bibliography
1(;25] Conroy, B.P., Sumner, M., Alexander, T., Application of encoderless vector control techniques in a medium performance induction motor drive. Proc. EPE 95, Sevilla, pg. 3.469-3.474 1<:26] Cordes, S., Nestler, J., Matrixconverter zur Speisung von drehzahlver
iinderlichen Drehfeldmaschinen. Proc . SPS/IPC/DRIVES 99, Nürnberg, pg.
531-540
I( :27] Cordeschi, G., Parasiliti, F., A variable structure approach for speed control
of field-oriented induction motor . Proc. EPE 91, Firenze, pg. 2-222
I( :28] Cornell E.P., Lipo, T.A., Modelling and design of controlled current induc
tion motor drive systems. IEEE Trans. Ind. Appl. 1977, pg. 321
1<:29] Coussens, P.J. Van den Bossche, A.P., Melkebeek, J .A., Nonlinear field ori
ented control in a rotor reference frame. Proc. EPE 95, Sevilla, pg. 1.820 1.825
II) I] Da Costa Branco, P.J., Stephan, R.M., A simple adaptive scheme for indirect
field orientation of an induction motor. Proc. EPE 91, Firenze, pg. 2-208
11)2] Dakhouche, K ., Roye, D., Digital vector control of induction machine using
a PWM inverter. Proc. EPE 91, Firenze, pg. 2-227
[I rI] Dallago, E., Sassone, G., Study and application of a digital speed control in
a DC brushless drive. Europ. Trans. on Electr. Power Engng. 1991, pg. 169
II )ii] Dallhammer, R., Gotthardt, K.-J ., The operation of microcomputers in rail
ways with the example of the H-Bahn. Microelectronics in Power Electronics
and Electrical Drives, Darmstadt 1982, ETG-Fachberichte 11, pg. 409
IIIS] Dannhauer, D ., Prüfstand für Werkzeugmaschinenantriebe mit einer per
manenterregten Synchronmaschine ais Belastungseinheit. Diss. TU Braun
schweig 1993
II )(i] Daum, D ., Unterdrückung von Oberschwingungen durch Pulsbreitensteue rung . ETZ-A, 1972, pg. 528
Daum, D ., Digitale Steuereinrichtung für Stromrichteranlagen. ETZ-A , 1973,
11)71 pg. 299
II )H I Daum, D ., Untersuchung eines Einphasenstromrichters mit nahezu si
IIllsfõrmigem Netzstrom und gut geglatteten Gleichgrofien. Diss. Univ.
Hudlum 1974
I\) l l I Ibvis, R .M., Melling, J., Quantitative comparison of commutation circuits
rnr bridge inverters. Proc. IEE 1977, pg. 237
[lHOI Ihvis, R .M., A variable speed drive that took a century and a halfto develop.
1':ledrieal Times, 1983
II )I II 1)(: Carli, A., Marola, G., Optimal pulse-width modulation for three-phase
voltage supply. Microelectronics in Power Electronics and Electrical Drives,
I);most.adt 1982, ETG-Fachberichte 11, pg. 197
rd [I) 12 1 De Carli, A, Steady state behaviour of converter fed systems. 3 IFAC
Sylllp. Control in Power Electronics and Electrical Drives. Lausanne, 1983,
Snrvey
II >1:1[ 1k Carli , A., Qualification of drives for motion control. Proc. EPE 91,
I,'irenze, pg. 2-589
1 1 )11\1 1><': Carli, A. , Onofri, 1., Design and implementation of a robust control strat
('g.Y for mechanical systems. Proc. Workshop on Advanced Motion Control
(AMC) 98, Coimbra, pg. 47-52
ii mi I)., I)Olldcer , R .W.A.A., Synthesis and digital implementation of adaptive li( :I
Iwd dl'l~ t> har compensation. Diss. Kath. U. Leuven, 1986
~ I )I (>[ I k I)Ollclwr , rt.w. Â. A., Fidd-oricnlcel <.;olll.rolkrs wi III rotor deep Imr C.()lll 11C ' 1\l 1 ;~ l.í"" cir(,lIil. ~ . 11,:1';[0: 1'rall~. I"d . Âppl. I~l!l :.!, pg . IO() ~
415
[D17] de Fornel , 8., Bach, J .L., Pietrzak-David, M. , Controllaw of an AC variable
[D18] [DI9] [D20)
[D21]
[D22]
~~ ~~
~~ [D2~ [D2~
~~ ~~
~~ [D31] [D32] [D33] [D34] [D35] [D36]
speed drive without mechanical sensors. Microelectronics in Power Electron
ics and Electrical Drives Darmstadt 1982, ETG-Fachberichte 11, pg. 329
Degner, M.W., Lorenz, R.D., Wide bandwidth flux, position and velocity estimation in AC machines at any speed (including zero) using multiple saliencies. Proc. EPE 97, Trondheim, pg. 1.536-1.541 de Jager, W .A.G., Tubbing, G.H., A vector oriented control strategy for a
4-quadrant line side converter. Proc. EPE 93, Brighton, pg. 5-213
de la Vallée, H., Grenier, D., Legat, J.-D., Labrique, F., A dedicated multi
processor architecture for AC motor cont.rol. Proc. IECON 98, Aachen, pg.
1632-1637
Delécluse, Ch., Grenier, D., A measurement method of the exact variation
of the self and mutual inductances of a buried magnet synchronous motor
and its application to the reduction of torque ripples. Proc. Workshop on
Advanced Motion Control (AMC) 98, Coimbra, pg. 191-197
Dell ' Aquila, A., Cupertino, F ., Salvatore, L., Stasi, S., Kalman filter estima
tors applied to robust control of induction motor drives. Proc. IECON 98,
Aachen, pg. 2257-2262
Deleroi, W., Winterling, M.W ., Commutation analysis of synchronous ma
chine drives. Proc. EPE 91, Firenze, pg. 4-572
Depenbrock, M., Einphasenstrornrichter mit sinusfórmigem Netzstrom und
gut geglatteten GleichgroJ3en. ETZ-A, 1973, pg. 466
Depenbrock , M.; Einphasenstrornrichter mit optimiertem Leistungsfaktor.
ETZ-A, 1974, pg. 360
Depenbrock, M., Direkte Selbstregelung (DSR) für hochdynamische Dreh
feldantrieb e mit Umrichterspeisung, ETZ-Archiv 1985, pg. 211
Depenbrock, M., Direct self control (DSC) of inverter-fed induction machine.
IEEE Trans. Power Electr . 1988, pg.420
Depenbrock, M., Hoffmann, F., Koch, S., Speed sensorless high performance control for traction drives. Proc. EPE 97, Trondheim, pg. 1.418-1.423 Depenbrock, M., Foerth, Ch., Koch, S., Speed sensorless control of induction motor at very low stator frequencies. Proc. EPE 99, Lausanne Depping, F., Voits, M., Automatic selection of control algorithms for an electrical drive with microcomputer-based speed control. 3 rd IFAC Symp.
Control in Power Electronics and Electrical Drives, Lausanne 1983, pg. 507
Dessaint, L.A., Saad, M., Hebert, B., AI-Haddad, K., An adaptive controller
for a direct-drive scara robot . IEEE Trans. Ind . Electr. 1992 , pg. 105
Dietrich, D ., Naunin, D., Closed loop speed control of an induction motor
by microcomputers. Microelectronics in Power Electronics and Electrical
Drives , Darmstadt 1982, ETG-Fachberichte 11, pg . 335
Dirr, R, Neuffer, I., Schlüter, W., Waldmann, H., Neuartige elektronische
Regeleinrichtungen für doppeltgespeiste Asynchronmotoren grofier Leistung.
Siemens Zeitschrift 1971, pg. 362
Dittrich, A., Hofmann, W ., Stoev, A., Thieme, A., Design and control of
a wind power station with double fed induction generator. Proc. EPE 97,
Trondheim, pg. 2.723-2.728
Dobrovski, I.A., Beitrage zum Entwurf und zur Dimensionierung von Thy
ristor-Gleichstrom-Umrichtern. Diss. TU Braunschweig 1976
Dodds, J .S., Vittek , J ., Mienkina, M., Implementation of a sensorless induc
liun lTI(,tor drive control system wi t h prescribed closed-Ioop rotor magnctic
IIII X IL llcI spe(!d clYll lunics. proe:. lõa) l ~ ~ 17 ) ' I 'l'''llClIll:illl , pg. 4.1l!.l2-1\.1\()7
'11G
Bibliography
Bibliography
Dote, Y., Stabilization of controlled current induction motor drive system via new nonlinear state observer. IEEE Trans. Ind. El. Contr. Instr. 1980, pg. 77 IIl:\HJ Dote, Y., Anbo, K ., Combined parameter and state estimation of controlled current induction motor drive system via stochastic nonlinear filtering tech niques. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1979, pg. 838 II J:\!)] Dote, Y., Manabe, T., Murakami, S., Microprocessor-based force control for manipulator using variable structure with sliding mode. 3"d IFAC Symp. Control in Power Electronics and Electrical Drives, Lausanne 1983, pg. 145 II 1~()1 Drozdowski, P., Field oriented control of the induction motor fed by a 3x3 matrix converter. Proc. EPE 99, Lausanne IIHII Du, T., Brdys, M.A ., Taufiq, J.A., Analytical parameter sensitivity expres sions for a field oriented induction motor drive . Prac. EPE 91, Firenze, pg. 3-623 11" '.!] Dunford, W .G., Dewan, S.G., The design of a control circuit for a two quadrant chopper based on the motorola 6800 microprocessor. IEEE Trans. Ind. Appl. 1980, pg. 495 II': I J Eckhardt, H ., Berechnung des quasistationii.ren Betriebs einer pulswechsel richtergespeisten Synchronmaschine kleiner Leistung. ETZ-Archiv 1982, pg. 321 11':21 Eder, E., Die Ungleichfórmigkeit der von einem dekadischen Frequenzteiler crzeugten Pulsfrequenzen. Regelungstechnik 1976, pg. 53 11·::q fi:dvardsen, P.A., Nestli, T.F., Nilsen, R., Kolstad, H., Steady state power now and efficiency optimizing analysis of a variable speed constant frequency generating system. Proc. EPE 97, Trondheim, pg. 2.691-2.694 I';ggert , B ., 1.5 MW wind power station with low AC line distortion using a iJ':." standard doubly-fed generator with field orientation control. Proc. EPE 97, Trolldheim, pg. 2.739-2.742 II.; ~ ,I I':ihd, .1 ., Ji:itten, R., Control of a 3-level switching inverter, feeding a three plo;~'i l' lIlachine, by a microprocessor. Microelectronics in Power Electronics 1\1,<1 1': lldri cal Drives, Darmstadt 1982, ETG-Fachberichte 11, pg. 217 1':i: ;( 'II :u:k, II., Hofmeister, H., Digitale Nachbildung von elektrischen Netzw 11 <1i[ ," '''' '11 IlIil Dioden und Thyristoren. AtE 1972, pg. 32 II.:'(I I':I!,.<:I' , Ir., Weil3 , M., Untersynchrone Stromrichterkaskade ais drehzahlregel 1"II'('r i\lItri eb für Kesselspeisepumpen. Siemens Zeitschrift 1968, pg. 308 1':Il'l,rillgham, L. , Wheeler, P., Clare, J ., Bi-directional switch commutation 1';11[ {(Ir 'llalrix converters. Proc. EPE 99, Lausanne 11';01 1':11<10 , '1'., Tajima, F., Okuda, H., lizuka, K., Kawaguchi, Y., Uzuhashi, 11., Oleada, Y., Microcomputer-controlled brushless motor without a shaft Illollllt e d position sensor. Proc. Int. Power Electr. Conf., Tokyo 1983, pg.
IIn7J
1
!): \~
II ': I () I 1':II(;I.lhardt, W., Schlingenheber und deren Antriebe an Fertigstrafien von
I\r, ~ itbaJl(lwal zwerken . Siemens Zeitschrift 1973, Beiheft Antriebs-technik
11': 111
1111.1 I'rozefhutomatisierung in Hütten-und Walzwerken, pg. 89 1 ': Hjl~ ti , P.N ., Ziogas, P.D. , Lindsay, J.F ., A current source PWM inverter lViL" illsL,tlltaneous current control capability. IEEE Trans. Ind. Appl. 1991, I )I~. ~ H'2
WI '21 II': I :q
1':lIjl 'li, P.N. , Ziogas, P.D., Lindsay, J.F., Rashid, M.H., A new current control "d"'lIIc ror AC motor drives. IEEE Trans. Ind . Appl. 1992, pg. 812 1.:rtlsL-Cn.t.hor, .I., J.)rl~h-,,(lhl variable Windellcrgie:wlagc rnit Glricltstro!O'l,w i :w lll'lIlll'(' is- l lmric.hkr Ilnd "dlw";!',I"IH(;
oplÍ11111 1l iu/'!III:III:lId"1I1
1l.(:p;lpr.
J)iRs
TU I1mllll
417
[E14J Ertugrul, N., Acamley, P.P., French, C.D., Real-time estimation of rotor position in PM motors during transient operation. Proc. EPE 93, Brighton, pg. 5-311 [E15] Eschenauer, H., Henning, G., Trondle, H.-P., Digitale Simulation des dy namischen Verhaltens und der Antriebsregelung von Grol3antennen. Siemens Zeitschrift 1974, pg 833 [E16] Espelage, P.M., Chiera, J.A.,Tumbull, F.G ., A wide range static inverter suitable for AC induction motor drives. IEEE Trans. Ind. Gen. Appl. 1969, pg. 438 [EI7] Ettlinger, G., Leitgeb. W., Poppinger, H., Simotron-Antriebe mittlerer Leis tung. Siemens Zeitschrift 1971, pg. 186 Fadei, M., de Fornel, B., Control laws of a synchronous machine fed by a [FI] PWM volt age source inverter. Proc. EPE 89, Aachen, pg. 859 Farrer, W., Miskin, J.D., Quasi-sine-wave fully regenerative inverter. Prac. [F2J IEE 1973, pg. 969 [F3J Favre, J .-P., Microprocessor-based speed control of a SCR-DC-motor. Mi croelectronics in Power Electronics and Electrical Drives, Darmstadt 1982, ETG-Fachberichte 11, pg.257 [F4] Fedor, P., Fedak, V., Bober, P., Timko, J., Field-oriented induction motor drive with indirect rotor flux sensing. Proc. EPE 91, Firenze, pg. 2-214 Fenker, O., Schumacher, W., Control of an induction motor without shaft [F5J encoder using the VECON chip. Proc. EPE 97, Trondheim, pg. 1.430-1.433 Fenker, O., Wirkungsgradoptimierte und drehgeberlose Regelung von Asyn [F6J chronmaschinen. Diss. TU Braunschweig 1998 Ferranti, M., Ferraris, L., Laner, E., Villata, F., Induction motor drives for [F7] traction application. Prac. EPE 93, Brighton, pg. 5-282 [F8] Ferraris, P., Fratta, A., Vagati, A., Villata, F., About the vector control of induction motors for special applications without speed sensor. Prac. Evo lution and modem Aspects of Induction machines, 1986, Torino, pg. 444 [F9J Ferretti, G., Magnani, G., Rocco, P., Online identification and compensation of torque disturbances in permanent magnet AC motors. Proc. IECON 98, Aachen, pg. 1521-1526 [F10J Fetz, J., Horstmann, D ., Comparison of different field oriented contrai meth ods for an induction motor fed by a PWM controlled inverter. Proc. EPE 89, Aachen, pg. 1079 [FIl] Fieger, K., Über die Anwendung der Abtasttheorie auf nichtlineare Regel kreise mit nichtstetigen Stellgliedern. Diss. TU Braunschweig 1967 [F12J Fieger, K.: Zum dynamischen Verhalten thyristorgespeister Gleichstrom Regelantriebe. ETZ-A, 1969, pg. 311 [F13J Fieger, K ., Middel, J., Vereinfachte Behandlung linearer Abtastregelkreise. Regelungstechnik 1967, pg. 445 [F14] Finlayson, P.T., Washburn, D .C., Cycloconverter controlled synchronous machine for load compensation on AC power systems. IEEE Trans. Ind. Appl. 1974, pg. 806 [F15] Fischer, J., Digitale Fahrkurvenrechner fur Positionieranlagen. AEG Mit teilungen 1976, pg. 269 [F16J Fischer, R., Geissing, H., Halbleiterstromrichter in kreisstromfreier Gegen parallclscha ltung für Umkehrantriebe. Siemens Zeitschrift 1964, pg. 240 [l"17J Flac,J., K , IToffmann, R., Mutschler, P., Direct mean torque control of an illdllcl.ioll 1I11l1.() r. Proc. EPB 97, TI'Ullcl\willl, pg. 3.672-3.677 [Io'I HI Vli il., ·, H., VI'II.c), II., Di \' l.hyriHt.nrW'fi p" iHtl' (o'iirdprm (L~chillC in Peldumkcbr liI· IIIIII.IIIII', '
1\1':<: Mil.lc,iJIIIII·.I·1I
1!l 7 :~)
PI'.. ·.~ r) /1
418
Bibliography
[F19] Flõter, W., Ripperger, H., Die TRANSVEKTOR-Regelung für den feldori
entierten Betrieb einer Asynchronmaschine. Siemens Zeitschrift 1971, pg.
761
[F20] Flügel, W., Drehzahlregelung der spannungsumrichtergespeisten Asynchron
maschine im Grund-und Feldschwachbereich. ETZ-Archiv 1982, pg. 143
[F21] Foerst, R.: Der Lõschwinkelregelkreis einer Hochspannungs-Gleichstrom
Übertragung mit symmetrischer Drehspannung. Diss. TH Darmstadt 1967
[F22] Fossard, A.J ., Clique, M., A general model for switched DC-DC converters
including filters. 2nd IFAC-Symp. Control in Power Electronics and Electri
cal Drives, Düsseldorf 1977, pg. 117
[F23] Frankl, G., Control modes for a variable speed drive with a converter-fed
synchronous machine. Proc. EPE 91, Firenze, pg. 4-592
[F24] Fratta, A., Vagati, A., Villata, F., Vector control of induction motors without
shaft transducers. Proc. IEEE-PESC 88, Kyoto, pg.839
[F25] Fratta, A., Grassi, R ., Vagati, A., High performance spindle drives adopting
IPM synchronous machines; theoretical goals and a practical solution. Proc.
EPE 91, Firenze, pg. 3- 476
[F26] Fratta, A., Vagati, A., Villata, F ., Extending the voltage saturated perfor
mance of DC brushless drive. Proc. EPE 91, Firenze, pg. 4- 134
[F27] Fratta, A., Vagati, A., Villata, F., On the evolution of AC machines for
spindle drive applications. IEEE Trans. Ind. Appl. 1992. pg. 1081
[F28] Freund, E. , Hoyer, H., Das Prinzip nichtlinearer Systementkopplung mit der
Anwendung auf Industrieroboter. Regelungstechnik 1980, pg. 80
[F29] Frick, A., von Westerholt, E., de Fornel, B., Nonlinear control of induction
motors via input- output decoupling. Europ. Trans. on Electr. Power Engng.
1994, pg. 261
[F30] Fritzsche, W., Genaue und schnelle Regelungen von Drehzahlen durch digi
tale Methoden. AEG Mitteilungen 19{30, pg. 419
[F31] Fujimaki, T., Ohniwa, K., Miyashita, O ., Simulation of the chopper con
trolled DC series motor. 2nd IFAC-Symp. Control in Power Electronics and
Electrical Drives, Düsseldorf 1977, pg. 609
[F32] Fujimoto, H., Kawamura, A., Perfect tracking digital motion control based
on two-degree-of-freedom multirate feedforward control. Proc. Workshop on
Advanced Motion Control (AMC) 98, Coimbra, pg. 322-327
[F33] Fujita, K., Sado, K., Instantaneous speed detection with parameter identifi
cation for AC servo systems. IEEE Trans. Ind. Appl. 1992, pg. 864
[F34] Fujiyama K., Tomizuka, M., Katayama, R ., Digital tracking controller desigu
for CD player using disturbance observer. Proc. Workshop on Advanced
Motion Control (AMC) 98, Coimbra, pg. 598-603
[F35] Fukunda, S., Itoh, Y., Nii, A., A microprocessor-controlled speed regula
tor for commutator-Iess motor drives. Proc. Int. Power Electr. Conference,
Tokyo 1983, pg. 938
[F36] Furuhashi, T., Sangwongwanich, S., Okuma, S., A position- and velocit..Y
sensorless control for brushless DC motors using an adaptive sliding modt:
observer. IEEE Trans. Ind. Electr. 1992, pg. 89
[G1] Gabriel, R., Leonhard, W ., Nordby, C., Microprocessor control of indllct.ioll
motors employing field coordinates. Proc. Int. Conf. on Electrical Va ri nhlc '
Speed Drives, IEE 1979, London , pg. 146
[G2] Gabriel, R., Leonhard, W ., Nordby, C., Rcr,elulI g der ~l.romri c ht.('q;,'s p (l i H t.c · 1I
Drehst.rom-Asynchronmaschine mit (:ilICJII M ilUYm·:chllc:r. B J:KC h111 ~{:; I.C' .. hll il\
197!!, pr~ . ;t7()
/< ::q <:n.h.. i, ·I, H.. , I,c 'olllm!''', W., No!'dhy, (: , l,'i,·lcl 01 j'·ld,nl.I·" (:0111.1'01 "r I~ tl l.ll."" " I"
1\<: II">!.,,,, " "i l'l'. l"i l'ro PI'OC "·: I/lClI'I I. 11 ': 1': 1'; ' 1'1'1'"11. III CI. I\p pl. IOHIJ , 1'1 '.. IHII
Bibliography [G4]
41!J
Gabriel, R., Leonhard, W., Nordby, C., Microprocessor control of tlI(:
converter-fed induction motor. Process Automation 1980, pg. 35
[G5] Gabriel, R., Feldorientierte Regelung einer Asynchronrnaschine mit einem
Mikrorechner. Diss. TU Braunschweig 1982
[G6] Gabriel, R., Leonhard, W., Microprocessor control of induction motor. Proc.
Int. Semiconductor Power Converter Conf. 1982, Orlando, pg. 385
[G7] Ganji, A.A., Lataire, Ph., Rotor time constant compensation of an inductioll
motor ln indirect vector controlled drives. Proc. EPE 95, Sevilla, pg. 1.431 1.436
[G8] Garces, L.J., Ein Verfahren zur Parameteranpassung bei der Drehzahlre- ,
gelung der umrichtergespeisten Kafiglaufermaschine. Diss. TH Darmstadt
1978
[G9] Garces, L.J., Parameter adaption for the speed controlled static AC drive:
with squirrel cage induction motor. IEEE Trans. Ind. Appl. 1980, pg. ln
[GlO] Garcia-Cerrada, A., Taufiq, J.A., Convergence of rotor flux estimation iII
field oriented control. Proc. IEE-Power Electronics and Variable SpC'/:cI
Drives, London 1988, pg. 291
[Gll] Garcia, G.O., Stephan, R.M., Watanabe, E .H. Comparing the indirect lidcl
oriented control with a scalar method. Proc. IECON 91, Kobe, pg. 475
[G12] Gastli, A., Matsui, N., Stator flux controlled V If PWM inverter with idc'"
tification of 1M parameters. IEEE Trans. Ind, Electr. 1992, pg. 334
[G13] Gebler, D ., Holtz, J., Identification and compensation of gear backlash wil.ll
out output position sensor in high-precision servo systems. Proc. IECON IH\,
Aachen, pg. 662-666
[G14] Gekeler, M, Jõtten, R., Control device for a pulsed 3-level illv<:/,(.c:1' k,'d ing a three-phase machine implemented by EPROMS and MSI digit.fl l ('" cuitry. Microelectronics in Power Electronics and Electrical Driv('H, I)" , '" stadt 1982, ETG-Fachberichte 11, pg. 207
[G1~ Gelman, V., High performance current source inverter drive. Proc. 1111., S,.""
conductor Power Conv. Conf. 1982, pg. 266
[G1~ Gelot, N.S., AIsina, P.J., A discrete model of induction motors ror I'C 'III"I.;III "
control applications. IEEE Trans. Ind. Electr. 1993, pg. 317
[G1~ Georgiou, G ., Le Pioufle, B., Nonlinear speed control of a syllclorollClIIH ri '"
vomotor with robustness. Proc. EPE 91, Firenze, pg. 3-42
[G1~ Gerecke, E ., New methods ofpower control with thyristors. SlIrvI ',Y, :1",( 11,'1\< :
Congress London 1966
[G1~ Geschke, B., Haspelanlage einer Warmband-Kontistraf3e. Sielllc'II!1:/.c·;I.'IC'i ,1'l 11.
1973, Beiheft Antriebstechnik und Proze13automatisierullg iII l'Iiil.I,'·'1I ,,".1
Walzwerken, pg. 97
[G20] Geyer, W ., Regelungstechnische Probleme bei der Regelllng ";IIC'1l (:1",,,,,
strommotors bei gleichzeitigem Eingriff in Anker-und FeldstrO\llkrc ·,,1. III 1': 11
ergieelektronik und geregelte elektrische Antriebe VDE-Buch fC1 iloc', VoI . I I
1962, pg. 472
[G21] Geyer, W ., Strõle, D., Entwicklungstendenzen bei geregelt.cn dnld.d u!'! ",,,
Antrieben. Siemens Zeitschrift 1965, pg. 490
[G 22] Gcyer, W ., Waller, S., DirekHührung numerisch gesteuerl,(:r W, 'rI";""J \
maschinen mit einem 1....:rt.ig ulIgslr:itrechllcr. Siemcm; Zeit!:ichrirl. 11170 , II",
hcft N urnerisdH: Stl~ \l C: rIln p;c' II , PI~ . II
le:'.:!) c:c wcc:kc", .I" IJlIl.cti'i ynchrulI" nf..tI ~II, rcllll ~ llrC:1lISi'ic ltn1(.lIn g< ·U fiil' 1I1·1)(':r.II
~20
Bibliography
Bibliography
IC25] Gõlz, G., Grumbrecht, P., Vmrichtergespeiste Synchronmaschinen. AEG Mitteilungen 1973, pg. 141 1<:26] Gõtz, F.-R., Synchronisierte Einzelantriebstechnik für Druckmaschinen. Proc. SPS/IPC/DRlVES 97, Nürnberg, pg. 390-399 1<: 27] Gosbell, V.J., Dalton, P.M., Current control of induction motors at low speed. IEEE Trans. Ind. Appl. 1992, pg. 482 1< :28] Grenier, D., Yala, S., Akhrif, O., Dessaint, L.-A., Direct torque control of PM motor with nonsinusoidal flux distribution using state-feedback linearization techniques . Proc. IECON 98, Aachen, pg. 1515-1520 IC2 U] Greubel, F., Huth, G ., Weiterentwicklung permanentmagnet-erregter Lin earmotoren. Proc. SPS/ IPC/DRIVES 99, Nürnberg, pg. 745-754 1<::10] Grieve, D.W., Cross, D M., Recent developments in the design and appli cation of large, high performance cycloconverter drive systems. Proc. IEE Power Electronics and Variable Speed Drives, London 1988, pg. 309 1< ::\ I] Grõtzbach, M., Analyse von Gleichstromstellerschaltungen im lückenden und nichtlückenden Betrieb. ETZ-Archiv, 1979, pg. 29 I( ::\2] Grofi H., Riedel, H. , Angepafite Vorschubantriebe für Werkzeugmaschinen. Siemens Zeitschrift 1976, pg. 30 1< ::U] Grotstollen, H., Zusammenhange und Kennwerte bei der Bemessung beschle unigungsoptimaler Antriebe. ETZ-A, 1974, pg. 663 1< ::\~] Grotstollen, H., Comparison of speed controlled DC drives with and without subordinate current control. 2 nd IFAC Symp., Control in Power Electronics and Electrical Drives, Dusseldorf 1977, pg. 583 1< ::I!í) Grotstollen, H., Pfaff, G., Bürstenloser Drehstrom-Servoantrieb mit Erre gung durch Dauermagnete. ETZ 1979, pg. 1382 I( ::11'1 Gruber , M ., Mõller-Nehring, P., Digitaler Geschwindigkeitsregler auf Mikro prozessorbasis im System MODULPAC C. Siemens Energietechnik 1979, pg. 117 I( ::1"7) Ciithlein, H., Die neue elektrische Lokomotive 120 der Deutschen Bundes !J'I.lIII in Drehstrom-Antriebstechnik. Elektrische Bahnen, 1979 II ::IKI Cllill ce, R .A., Lyden, C., A novel and accurate estimation method of the sl.(·1\("" v"cl,II" "IC"lllllI.l.i"ll . II':I':!': ' I'rallll . .1,,11 1\ 1'1'1
1 \l 1I'.~ , 1'1'. ' IIH I.
[H4]
421
Habõck, A., Speisung von Synchronmaschinen über Umrichter mit einge pragtem Strom im Zwischenkreis. VDE Fachtagung Elektronik 1969, pg. 163 [H5] Habõck, A., Kõllensperger, D., Stand und Entwicklung des Stromrichtermo torso Siemens Zeitschrift 1971, pg. 177 [H6] Hackstein, D ., Time optimal 3-dimensional trajectory control. 3 rd IFAC Symp. Control in Power Electronics and Electrical Drives, Lausanne 1983, pg. 553 [H7] Hadzimejlic, N., Sabanovic, A., Behlilovic, N., Electrical vehicle main drive control system. Proc. EPE 99, Lausanne [H8) Harer, H., Horn, V., Investigation of a doubly-fed induction generator using a simple cycloconverter. Proc. EPE 91, Firenze, pg. 2-254 [H9] Hagiwara, A., Tsuchiya, J., Shimizu, T., Kimura, G., Watanabe, 1., Naniwa, K., Speed sensorIess field-oriented control based on phase difference. Proc. IECON 98, Aachen, pg. 1014-1017 [H10] Halász, S., Optimal control of voltage source inverters supplying induction motors . 2 nd IFAC Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 379 [H 11] Halász, S., Zaharov . A., Analysis of two-phase PWM technique in inverter AC drives . Proc. EPE 99, Lausanne [HI2] Hannakam, L., Übergangsverhalten des Drehstrom-Schleifringlaufermotors. Regelungstechnik 1959, pg. 393, 421 [H13] Hannakam, L., Digitales Universalprogramm für Schaltvorgange an Dreh strom-Asynchronmaschinen. ETZ-A, 1961 , pg. 493 [HI4] Hapiot, J.C., Pietrzak-David, M., Bach, J.L., Regulation de vitesse d'une machine asynchrone alimentée en courant sans capteur de vitesse. Conumel 83, Toulouse, pg. IV-12 [H15] Harashima, F., Yanase, T., State space analysis of AC motors fed by thyris tor inverters. 18t IFAC Symp ., Control in Power Electronics and Electrical Drives, Düsseldorf 1974, Vol. 1, pg. 149 [H16] Harashima, F ., Naitoh, H., Dynamic performance of converter-fed syn chronous motors. 2 nd IFAC-Symp. Control in Power Electronics and Elec trical Drives , Düsseldorf 1977, pg. 369 [H17) Harashima, F., Hayashi, H. , Dynamic performance of current source inverter fed induction motors. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1978, pg. 904 [H18] Harashima, F ., Naitoh, H ., Haneyoshi, T ., Dynamic performance of a self controlled synchronous motor fed by current source inverters. IEEE Trans. Ind. Appl. 1979, pg. 36 [HI9] HarIey, R.G. , Wishart, M .T ., Diana, G., Teaching field oriented control of VSI and CSI fed asynchronous and permanent magnet synchronous machines using a programmable high speed controller. Proc. EPE 91, Firenze, pg. 2 556 [H20) Harms, K, Beitrage zur Regelung der Drehstrom-Asynchronmaschine unter Berücksichtigung veranderIicher Maschinenparameter. Diss. TV Braun schweig 1988 [H21] Harnefors, L., Taube, P. , Nee, H-P., An improved method for sensorless adaptive control of permanent-magrwt synchronous motors. Prac. EPE 97, Trondheim, p g. 4.541-4.546 [H22] Hamdors , L., Tnstabiliiy phenollll 'llll iII H('nsorles5 control of induction mo tors . 1'l'Oe. F:PE 99, Lausnnll(' 111:nl 'I la r~ , W ., Si ('. ht:llphll1'l il~( ·r ~ yll C "" ' ''I , " r '' .l ll' illil. 1'1I!>i w('clJcJrichl.er iII hochiu" 1,'·Wi,·"I., ', 1I IL Ilw,·i:w . 1':'1'<: 1>':1.<'111" ",11'11 1.... N., ;~ ;>', I!n,~\, VIlI': V" r\;l(', Ikl'lill, ( 111', '111 '" ('1,
t122
Bibliography Bibliography
1"24] Hartramph, R., Schwinkothe, W., Elektrodynamische Direktantriebe mit in
tegriertem WegmeBsystem. Proc. SPSjIPCjDRIVES 98, Nümberg, pg. 700 701
1"25J Hashimoto, S., Tanaka, K ., Miki, I., An autotuning method for vector con
trolled induction motor drives considering stator core losses. Proc. EPE 99,
Lausanne
11126] Hasse, K, Zur Dynamik drehzahlgeregelter Antriebe mit stromrichterge
speisten Asynchron-KurzschluBlãufermaschinen. Diss. TH Darmstadt 1969
I" 271 Hasse, K., Drehzahlregelverfahren für schnelle Umrichterantriebe mit strom
richtergespeisten Asynchron-KurzschluBlaufermotoren. Regelungstechnik
1972, pg. 60
1"28] Hattori, M., Ishikawa,T ., Taniguchi, M., Ohnishi, T., Kurata, Y., Kamiyama,
K ., Imagawa, K ., Takahashi, J., Sukegawa, T., Saito, K.: An experience with
cycloconverter-fed 1M drives for a modem tandem cold mil!. Proc. EPE 91,
Firenze, pg. 2-260
1"29J Healey, R.C ., The implementation of a vector control scheme based on an advanced motor mode!. Proc. EPE 95 , Sevilla, pg. 3.017-3.022 /I !:lO] Heinemann, C., Comparison of several control schemes for AC induction
motors under steady state and dynamic conditions. Proc. EPE 89, Aachen,
pg. 843
[II:! 1] Heinemann, G., Selbsteinstellende, feldorientierte Regelung für einen asyn
chronen Drehstromantrieb. Diss. TU Braunschweig 1992
li 1:!2] Heinemann, C., Leonhard, W., Self-tuning field-orientated control of an in
duction motor drive. Proc. Int. Power Electr. Conf. 90, Tokyo, pg. 465
1":\3J Heinemann, C., Papiernik, W., Hochdynamische Vorschubantriebe mit Lin earmotoren. VDI-Z Special Antriebstechnik, pg. 26-33, Springer Verlag, April 1998 II U1] Heinze, K., Tappeiner, H., Weibelzahl, M., Pulswechselrichter zur Drehzahl
steuerung von Asynchronmaschimen. Siemens Zeitschrift 1971, pg. 154
I":!:;] Heintze, K, Wagner, R., Elektronische Cleichstromsteller zur Ceschwin
digkeitssteuerung von aus Fahrleitungen gespeisten Cleichstrom-Triebfahr
ze ugen. ETZ-A, 1966, pg. 165
II I:!lij I leller, M., Die doppelt-gespeiste Drehstrommaschine für drehzahlvariable
l'umpspeicherkraftwerke. Diss. TU Braunschweig 1998
I"T/I lIeJler, M., Schumacher, W., Stability analysis of doubly-fed induction ma
chines in stator flux reference frame. Proc. EPE 97, Trondheim, pg . 2.707 2.710 I" :m] Hemmingsson, M., Simonsson, B., Alaküla, M., Olsson, G., Control and design interaction in hybrid vehicles - Benefits of using power electronics. Proc. EPE 97, Trondheim, pg. 1.566-1.569 jI UU] Hengsberger, A., Putz, U., Vetters, L., Thyristor-Stromrichter für Bahnmo
toren. AEC Mitteilungen 1964, pg. 435
1"1 U] Henneberger, C., Dynamic behaviour and current control methods of brush
lcss De motors. Proc. EPE 89, Aachen, pg. 1531
I" t1 Ij II en neberger, C., Brunsbach, B. J., Klepsch, Th., Field oriented control of
syJlchronous and asynchronous drives without mechanical sensors using a
I{ Illman-filter. Proe. EPE 91, Firenze, pg. 3-664
III ti:?] IlcJlllclwq.;cr, C., Lutter, T. , Brushless DC-motor with digital state con
troll.. r. l'roe . EPE 91, FireJl?;e, pg. 4-104
[l1t1 :q ILI'rU pill, 1\., '·:lltwiddlJllgst.elldcII:t.t:/1 IH ~i (\auddidowregdllllgen Wr KalI.·
lVill·/'w.·,rlu· . /11 ';( : M il.t."illlllg\'1\ I!)'7ü , pg . :!X
423
[H44] Heumann, K ., Jordan, K.C., Das Verhalten des Kã.figlaufermotors bei
veranderlicher Speisefrequenz und Stromregelung. AEC Mitteilungen 1964,
pg. 107
[H45] Hill, W.A. , Ho, E .Y.Y., Neuzil, I.J ., Dynamic behavior of cycIoconverter
systems. IEEE Trans. Ind. App!. 1991, pg. 750
[H46) Hillenbrand, F ., A method for determining the speed and rotor f1ux of the
asynchronous machine by measuring the terminal quantities only. 3rd IFAC
Symp. Control in Power Electronics and Electrical Drives Lausanne 1983,
pg. 55
[H47J Hoscheler, B., Erhohung der Cenauigkeit bei WegmeBsystemen durch selb
stlernende Komponenten systematischer Fehler. Proc. SPSjlPCjDRIVES
99, Nürnberg, pg. 617-626
[H48J Hovermann, M., Orlik, B., Schumacher, U., Field oriented control of induc tion motor without speed sensor. Proc. EPE 97, Trondheim, pg. 4.512-4.517 [H49] Hovermann, M., Orlik, B., Schumacher, U., Schümann, U., Operation of speed sensorless induction motors using open loop control at low frequency.
Proc. EPE 99, Lausanne
[H50J Hofer, K, Digital speed control by adaptive computation of firing angle
using microprocessors. Microelectronics in Power Electronics and Electrical
Drives, Darmstadt 1982, ETC-Fachberichte 11, pg. 295
[H51J Hoffmann, A H., Brushless synchronous motors for large industrial drives.
IEEE Trans. Ind. Cen. App!. 1969, pg. 158
[H52] Hoffmann, E., Schaufele, R., Einsatz einer regelbaren Speicherpumpe in der
Netzregelung, Elektrizitatswirtschaft, Heft 25, 1993
[H53] Hoffmann, F., Koch, S., Steady state analysis of speed sensorless control of
induction machines. Proc. IECON 98, Aachen, pg. 1626-1631
[H54] Hofmann, W., A suboptimal predictive current controller for VSI-PWM in
verters. Proc. EPE 91, Firenze, pg. 3-316
[H55) Hofstetter, M., Umrichter mit quasiresonant gepulstem Cleichspannungszwi
schenkreis für ein Drehstromantriebssystem. Archiv für Elektrotechnik,
1993, pg. 309
[H56) Hoft, R.C., Casteel, J.B., Power electronic circuit analysis techniques, sur
nd vey. 2 IFAC Symp., Control in Power Electronics and Electrical Drives,
Düsseldorf 1977, pg. 987
[H57J Hoft, R.G., Singh, D., Microcomputer-controlled single phase cycIoconverter.
IEEE Trans. Ind. Electr. and Contr. Instr. 1977, pg. 226
[H58] HolI, E., Neumann, C., Piepenbreier, B., Tolle, H .-J., Water-cooled inverter
for synchronous and asynchronous electric vehicle drives. Proc. EPE 93,
Brighton, pg. 5-289
[H59] Holliday, D., Creen, T .C., Williams, B.W., On-line parameter measurement
of induction machines. Proc. IECON 93, Maui, pg. 992
[H60] Holliday, D., Fletcher, J .E., Williams, B.W., Non-invasive rotor position and speed sensing of asynchronous motors. Proc. EPE 95, Sevilla, pg. 1.333-1.337 [H61] Holtz, J., Ein neues Zündsteuerverfahren für Stromrichter am schwachen Netz. ETZ-A, 1970, pg. 345
[H62) Holtz, J ., Kraftkomponenten und deren betriebliche Steuerung beim eisen
losen Synchronlinearmotor. ETZ-A, 1975, pg. 396
[H63] Holtz, J ., Schwellenberg, U., A new fast-response current control scheme for
line controlled converters. Proc. Int. Semiconductor Power Converter Conf.
1!IH:!, j>g. 175
fi 1(;t1 I . 11011.'1., .I ., Sl.nd I.Idd, S., /I pn ~d if:l.iv(· ("(JJII.rolh·r ror the ~tator cllrrent vector III" /I( : IIl lI dlill" " r.·d rrOlIl IL IIwíLc! ... d vIIII. IL /(" NOIIfT.':. Prol'.. lnt. . 1)()wI ~ r 1':IITl.ri('. ( '. ,(01, '''' ''' ''', 'I'"Jty .. IOH:!, PH . I fim,
~:.11
Bibliography
(H65] Holtz, J., Bube, E., Field-oriented asynchronous pulse-width modulation for high- performance AC machine drives operating at low switching frequency. IEEE 'l'rans. Ind. Appl. 1991, pg. 574 (H66] Holtz, J., Thimm, T., Identification of the machine parameters in a vector controlled induction motor drive. IEEE Trans. Ind. Appl. 1991, pg. 1111 (H67] Holtz, J., Pulsewidth modulation- a survey. IEEE Trans. Ind. Electr. 1992, pg.410 (H68] Holtz, J., Beyer, B., Optimal synchronous pulse width modulation with a trajectory- tracking scheme for high dynamic performance. IEEE Trans. Ind. Appl. 1993, pg. 1098 (H69] Holtz, J ., Speed estimation and sensorless control of AC drives . Proc. IECON 93, Maui, pg. 649 (H70] Holtz, J ., Beyer, B., Optimal pulsewidth modulation for AC servos and low cost industrial drives. IEEE Trans . Ind. Appl. 1994, pg. 1039 (H71] Holtz, J., Beyer, B., The trajectory tracking approach- a new method for minimum distortion PWM in dynamic high-power drives. IEEE Trans. Ind. Appl. 1994, pg. 1048 [H72] Holtz, J., Sensorless position control of induction motors - an emerging tech nology. Proc. AMC 98, Coimbra, pg. 1-14 [H73] Holtz, J., Sensorless position control- an emerging technology. Proc. IECON 98, Aachen, pg. 1.1-1.12 [H74] Hombu, M., Ueda, S., Ueda, A ., A current source inverter with sinusoidal inputs and outputs . IEEE Trans. Ind. Appl. 1987, pg. 247 [H75] Hombu, M., Futami, M., Miyazaki, S., Kubota, Y., Bando, A., Azusawa, N ., Harmonics analysis on a slip-power recovery system fed by a DC link GTO converter. Proc. EPE 95, Sevilla, pg. 3-239 [H76] Hopfensberger, B., Atkinson, D.J ., Lakin, R.A., Application ofvector control to the cascaded induction machine for wind power generation schemes. Proc. EPE 97 Trondheim, pg. 2.700-2.706 [H77] Hori, Y., Umeno, T., Uchida, T., Konno, Y., An instantaneous speed ob server for high performance control of DC servomotor using DSP and low resolution shaft encoder. Proc. EPE 91, Firenze, pg. 3-647 [H78] Hosono, L, Katsuki, K ., Yano, M., Static converter starting of large syn chronous motors. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1976, pg. 536 [H79] Huang, H ., Davat, B., Lajoie-Mazenc, M., Study of control strategies and regulation in cycJoconverter-fed synchronous machines. Proc. EPE 89, Aachen, pg. 871 [H80] Huang, L., Tadokoro, Y., Matsuse, K., Deadbeat flux levei control of direct field- oriented high-horsepower induction motor using adaptive rotor flux observer. IEEE Trans. Ind. Appl. 1994, pg. 954 [H81] Hugel, J., Die Berechnung von Stromrichter-Regelkreisen. AtE 1970, pg. 224 [H82] Huisman, H., Zero ripple torque control in brushless DC motors: a straight forward approach. Proc. EPE 95, Sevilla, pg. 3.851-3.856 [H83] Huppunen, J ., Pyrhõnen, J., Sine wave filter for PWM-inverter driven medium-speed Ci 30 000 rpm) solid-rotor induction motors. Proc. EPE 99, Lausanne [H84] Hussels, P., Mehne, M., Hentschel, F., Synthesis of cycJoconverter and cur rent source inverter- Presentation of a new control strategy. Proc. EPE 9:1, Brighton, pg. 5- 33 lnaha, n., Hirasawa, K., Ando, T. , Hornbu, M., N;dmz
BibliOI\fHpl,y
[12]
[13] [14] [15] [16] [17]
[Jl] [J2]
[J3] [J4] [J5] [J6] [J7]
[J8] [J9] [J10]
[Jll] [JI2]
[J13]
425
Irisa, T., Takata, S., Ueda, R., Sonoda, 'I '. , M" .. hi",J!,i, T., A novel approach on parameter self-tuning method in AC /{(! rvo systems. 3 rd IFAC Symp. Control in Power Electronics and Electrical Uri VI:S, L
126
Bibliography
Bibliography
[.114] Ji, J.K., Lee, D .C., Sul, S.K., LQG based speed controller for torsional vi bration suppression in two mass motor drive system. Proc. IECON 93, Maui, pg. 1157 [J15] Jõnsson, R., Method and apparatus for controlling an AC induction motor by indirect measurement of the airgap voltage, PCT Patent Application No. PCT/SE 91/00086, International filing date 8.2.91. European Patent 0515469, Issued 1994. US-Patent 5 294 876, Issued 1994 1.116] Jõnsson, R., Leonhard, W., Control of an induction motor without a me chanical sensor, based on the principie of "natural field orientation" (NFO). Proc. IPEC Yokohama, 1995, pg. 303 1.117] Jõtten, R., French Patent 1.331.612, 1961 111.8] Jõtten, R, Regelungstechnische Probleme bei elastischer Verbindung zwis
chen Motor und Arbeitsmaschine. ln Energieelektronik und geregelte elek
trische Antriebe, VDE- Buchreihe, Vol. 11, 1962, pg. 446
1.1 L9] Jõtten, R ., Die Berechnung einfach und mehrfach integrierender Regelkreise
der Antriebstechnik. AEG Mitteilungen 1962, pg. 219 [.120] Jõtten, R ., Signalverarbeitung für die Regelung umrichtergespeister Dreh feldmaschinen. 18t IFAC Symp ., Control in Power Electronics and Electrical Drives, Düsseldorf 1974, Vol. 1, pg. 681 1.121] Jõtten, R., Mader, G., Control methods for good dynamic performance in duction motor drives based on current and voltage as measured quantities. Proc. Int. Semiconductor Power Q,onverter Conf. 1982, pg. 397 [.122] Jõtten, R, Zimmermann, P., Sub-I.nd super-synchronous cascade with cur rent source inverter. r d IFAC-SyJ:p. Control in Power Electronics and Elec trical Drives, Lausanne 1983, pg. 337 1·I:n] Jõtten, R., Kehl, C., A fast space-vector control for a three levei voltage source inverter. Proc. EPE 91, Firenze, pg. 2-70 [.I21j] Jones, C.L, van Nice, R., Frequenzgeber und -teiler mit verlinderlicher Fre quenz. German Patent 1.100.084, 1958 1.l 2 ~)J Jones, R., Wymeersch, W., Lataire, Ph., Vector controlled drives in a steel processing application. EPE 93, Brighton, pg. 5-508 1. 1~~(iJ .Jurer, M., Indirect sensors for electric drives . Proc. EPE 95, Sevilla, pg. 1..836-1.841
1
1 1\ II ruisée par microprocesseurs d'une machine synchrone autopilotée. Conumel 8:1, Toulouse, pg. IV-19 IIOJ Kahlen, H., Maggetto, G., Electric and hybrid vehicles. Proc. EPE 97 Trondheim , pg. 1.030-1.054
[I\:\j Kamiyama, K., Azusawa, N., Miyahara, Y., Ohmae, T., Kivisawa, T., Micro
processo r controlled fast response speed regulator for thyristorised reversible regenerative DC-motor drives. IEEE Ind. Electr. and Control Instr. Ann. Meet. Rec. 1978, pg. 216 11(IjJ Kamiyama, K., Nagasaki, H ., Saito, K., Matsuka, S., Shiraishi, H., Ohmae, T ., Microprocessor-based high-speed internal controllers for industrial drives. Microelectronics in Power Electronics and Electrical Drives, Darm $ladt 1982, ETG-Fachberichte 11 \1<;) I Knu'l.aki, T ., Yanagida, K., Yasouku, L, Inverter control system for drivinl~ illdnctio!l'\l1oiors in rapid transit cars llsing high-powered GTO thyristorR. I' !'Oe:. Iut. SI"llicollljllçlor Power Converter Cnll r. 1982, PI~ · 145 II{IIJ 1\ l\ ~ H. , N., V"'~t.!lll l d H\ II., A Hc n ~orlc~H po:;ilioll çoutrol I>y sulicut.. polc brwdl Io ,tlfl 1)(: Ill"t,,, r. Pr"... . 1':1'(0; lI7 , 'Ii'o!ldhl 'illl, PI/;. ·1 .!o2!)-Ij .f.:JIj 11\ 71 li " ", Y.'1'. , I li II , ( : .11., {\ 1I 1l1 .v HiM IUld d rrdl\ 11 ,,/' IIIII ·r "l'rnl·I 'MsOr- llII./lc d vI'.c(."r '''111,11,11, ,, 1 IlId"I ·I. I\>1I 11I"I."r d r l Vl'rl . 11':1':1·: '1 'lI l1l1 l. IlId. 1':11 '1'1.1' . IH\l'!. , pl'..10
427
[K8] Kappius. F., Zur Drehzahlsteuerung von Drehstrom-Kurzschlul3laufermoto ren mit einer gleichstromgespeisten Stromrichteranordnung in sechspulsiger Mittelpunktschaltung. Diss. TH Braunschweig 1965 [K9] Kastner, G., Pfaff, G., Sack, L., Design and performance of an acceleration drive with permanent magnets. IEEE Trans. Ind. Appl. 1991, pg. 908 [K10] Kavanagh, R.C., Improved analysis and design techniques for digital tachom etry. Diss. Univ. College Cork 1998 [K11] Kavanagh, R.C., Murphy, J.M.D., The effects of quantization noise and sen sor non-ideality on digital-differentiator-based velocity measurement. IEEE Trans. Instrum. Meas. 1998, pg. 1457-1463 [K12] Kawamura, A., Yokoyama, T., Comparison offive control methods for digital feedback controlled PWM inverters. Proc. EPE 91, Firenze, pg. 2-35 [K13] Kazmierkowski, M.P., Kõpke, H.-J., Comparison of dynamic behaviour of frequency converter fed induction machine drives. 3 rd IFAC Symp. Conirol in Power Electronics and Electrical Drives Lausanne 1983, pg. 313 [K14] Kazerani, M., Ooi, B.T., Direct AC/AC matrix converter based on thre,,~ phase voltage- source converter modules. Proc. IECON 93, Maui, pg.812 [K15] Kazmierkowski, M.P., Kasprowicz, A., Improved torque and flux vector ÇOIl trol of PWM inverter-fed induction motor drive. Proc. EPE 93, Bright.oll, pg.5-115 [K16] Kazuno, H., A wide range speed control of an induction motor with ::;i a t.íc
Scherbius and Kramer systems. J. Electr. Engg. Japan 1969, pg. 10
[K17] Kelber, Ch.R., Aktive Dampfung der doppelt-gespeisten DrehstrollllllitS
chine. Diss. TU Braunschweig, 2000 [K18] Kennel, R., Schriider, D., Predictive control strategy for converters. 3,·,1 IJo'AC Symp. Control in Power Electronics and Electrical Drives, Lausanrw I !lln , pg. 415 [K19] Kennel , R., Kobs, G., Weber, R., Digital control of industrial servo driV('!) for machine tools. Proc. EPE 99, Lausanne [K20] Kerkman, R .J., Rowan, T.M., Leggate, D., Indirect field-oriented COlll.r,,1 I.. r an induction motor in the field-weakening region. IEEE Trans. Iud. Al'pl. 1992, pg. 850 [K21] Kerkman, R.J., Leggate, D., Seibel, B.J., Rowan, T.M., An overmodulnl.ion
strategy for PWM voltage inverters. Proc. IECON 93, Maui, pg. l:.!J~.
[K22] Kessler, C., Über die Vorausberechnung optimal abgestimmter Regd l\f(~i~c .
Regelungstechnik 1954: pg. 274, 1955: pg. 11, 40 [K23] Kessler, C., Das symmetrische Optimum. Regelungstechnik 1958, pg. :Wf), 432 [K24] Kessler, C., Meinhardt, W., Neuffer, L, Rube, G., Die Gleichstfl)Jllolliir dermaschine mit Siemens-TRANSIDYN-Regelung. Siemens Zeitschrif'i I !)!jH , pg. 555 [K25] Kessler, C., Ein Beitrag zur Theorie mehrschleifiger Regelungen. R egelullg stechnik 1960, pg. 261 [K26] Kessler, G., Das zeitliche Verhalten einer kontinuierlichen elastischen Bahn zwischen aufeinanderfolgenden Walzenpaaren. Regelungstechnik 1960: pg. 436, 1961: pg. 154 [K27] K essler, G., Digitale Regelung der Relation zweier Drehzahlen. ETZ-A, 1961, pg. 574 [K28] Kessler, G ., Digitale Regelung der Drehzahl und der Relation zweipr I)ndl~,ilhl l 'll. lu J)igital( ~ SigulLlvemrlwij,llllg iII der Rcgclungstechnik, VDI': IIllchreill (', V,,1. H, líJ62, PI/;. I !)',! c
[1<:21) I l': I'wdl'l',
(:., 1':I"I·l.ríclIl drivI' nylll.l-IIII I ror I' n p l\r · , rllbl,,' r- ILllIl plrl.lltil 'K, illdll:' 11o' lu~rI( · l n I !I'/ n
t, I"fl, III\(' VI ' Y III{\(: (:"11/'''''1',1)" .. ,
tl28
Bibliography
[K30) Khaburi, D.A., Tabar, F.M ., Comsa, A., Sargos, F.M., A DSP-based con troller for sensorless control of synchronous machines. Proc. EPE 97, Trond heim, pg. 4.547-4.552 [K31) Khambadkone, A., Holtz, J., Low switching frequency high power inverter
drive based on field oriented pulsewidth modulation. Proc. EPE 91, Firenze,
pg.4-672
[K32) Khanniche, M.S., Belaroussi, M., Sethuraman, S.K., An algorithm for gen
erating optimised PWM for real time micro control applications. Proc. IEE
Power Electronics and Variable Speed Drives, London 1988, pg. 386
1[(33) Kiel, E., Schumacher, W., Gabriel, R., PWM Gate-array for AC drives. Proc.
EPE 87, Grenoble, pg. 653
[[(34) Kiel, E., Schumacher, W., Ehrenberg, J., Letas, H. H., Schrader-Hausmann,
U., High performance digital control of uninterruptible power supply (UPS)
using an application specific integrated circuit (ASIC). Proc. EPE 91,
Firenze, pg. 3-174
[K35] Kiel, E., Anwendungsspezifische Schaltkreise in der Drehstrom-Antriebs
technik. Diss. TU Braunschweig, 1994
[K36J Kiel, E., Schumacher, W., Der VeCon-Prozessor, Signalverarbeitung für
Antriebe auf einem Chip. Proc. SPSjIPCjDRIVES 94, Stuttgart, pg. 533 539
[K37J Kiel, E., Schumacher, Veconi A high-performance single-chip-servovontroller
for AC drives. IPEC 95 Yokohama, 1995, pg. 1289
[K38J Kiel, E ., Schumacher, W., VECON: High-performance digital control of AC
drives by one-chip servO controller. Proc. EPE 95, Sevilla, pg. 3.005-3.010
[K39J Kiel, E., Loy, T., Der digitale Antriebsregler auf dem Weg vom ein
fachen Aktor zur dezentralen intelligenten Bewegungssteuerung. Proceedings
SPSjIPCjDRIVES 98, Nürnberg, pg. 476-487
[[(tiO] Kielgas, H., Nill, R, Converter propulsion systems with three phase induction
motors for electrical traction. Proc. lnt. Semiconductor Power Converter
Conf. Rec. 1977, pg. 304
1\ ti IJ Kim, G.S., Ha, LJ., Ko, M.S., Control of induction motors for both high
dynamic performance and high power efficiency. IEEE Trans. Ind. Electr.
1992, pg. 323
11\ 1:l1 IGm, G.S., Kim, J.Y., Choi, J.Y., Choy, L, Recursive rotor resistance adap
Lation algorithm for induction motor control. Proc. IECON 98, Aachen, pg.
883-887
11\ tI :1] Kim, H.Y ., Shin, M.H., Hyun, D.S., Improved vector control of an induction
motor with on-line tuning of its parameters. Proc. IECON 98, Aachen, pg.
854-858
II
ers with circulating-current mode, used in four-quadrant DC magnet power
supplies. IEEE Trans. lnd. Appl. 1991, pg. 1026
IJ( ti 5] Kim, T.W., Kawamura, A., Slip frequency estimation for sensorless low
speed control of induction motor. Proc. Workshop on Advanced Motion Con
irol (AMC) 96, Tsu City, pg. 156-161
[IWi] Kim, Y.H., Kim, S.S., Hong, I.P., Speed sensorless vector control of high
speed induction motors using intelligent control algorithm. Proc. IECON 98,
AacheJl, pg. 888-892
[1< tn j Killl , Y.lL , Sul, S.K., Park, M.n., Speed sensorless vector control of induc tioll lIIoLor Il ~ ill~ extended Kalman filter. IEEE Tralls. lnd. Appl. 1991, PI,' 1
t
l n!>
[1\ .11'11
1\ illllill"'III"
1) ..1 ., i\,·;tl'loIny, 1'.1'" .Iack, i\ .e , i\ li illt"wat,'d cin :llit ('(llll,roll",.
rIO,> "I'II :: ld' ·II" 1)(: ,h·;v,',l. 1'1' 0 1:. l'a'I ': !lI J FI .."n/,,· , 1'1\ ti I I I
Bibliography
429
[K49] Kinniment, D .J., Kappos, E., Acarnley, P.P., Experience of the use of ASIC
methods in a motor control application. Proc. EPE 93, Brighton, pg. 458
[K50J Kirlin, R.L., Trzynadlowski, A.M., Bech, M.M., B1aabjerg, F., Pedersen, J .K., Analysis of spectraI effects of random PWM strategies for voltage source inverters. Proc . EPE 97 Trondheim, pg. 1.146-1.151 [K51J Kisck, D.O., Bucurenciu, S., Bucurenciu, M., Sirbu, G., Kisck, M., Real-time parameter estimation of vector-control induction motor fed by CR-PWM extended Kalman filter. Proc. EPE 99, Lausanne [K52J Klaassen, H ., Selbsteinstellende, feldorientierte Regelung einer Asynchron
maschine und geberlose Drehzahlregelung, Diss. TU Braunschweig, 1996
[K53J Klaes, N.R.,Identifikationsverfahren für die betriebspunktabhangigen Para
meter einer wechselrichtergespeispen Induktionsmaschine. Diss. Univ. Bo
chum, 1992
[K 54] Klaes, N.R., On-Iine tuning of the rotor resistance in an inverter-fed in
duction machine with direct-self-control. Europ. Trans. on Electr. Power
Engineering. 1994, pg. 5
[K55J Klautschek, H., Das Verhalten der lnduktionsmaschine bei Speisung über
Strom-Zwischenkreisurnrichter. ETZ 1974, pg. 283
[K56J Klautschek, H., Asynchronmaschinenantriebe mit Strom-Zwischenkreisum
richtern. Siemens Zeitschrift 1976, pg. 23
[K57J Klein, G, Stebel, H., Steuerung und Regelung von Hochlaufscheren und sagen. BBC Nachrichten 1964, pg. 200
[K58J Kleinsorge, N., Putz, U., Stemmler, H., Large adjustable speed AC drives.
Proc. EPE 87, Grenoble, pg. 491
[K59J Kliman, G.B ., Plunkett, A.B., Development of a modulation strategy for a
PWM inverter drive. IEEE Trans. Ind. Appl. 1979, pg. 72
[K60J Klug, R .- D., Effects and correction of switching dead times in 3-phase PWM
inverter drives. Proc. EPE 89, Aachen, pg. 1261
[K61J Kobayashi, H., Endo, S., Kobayashi, S., Kempf, C.J., Robust digital tracking
controller for high speed positioning systems - a new design approach and
implementation techniques. Proc , Workshop on Advanced Motion Control
(AMC) 96, Tsu City, pg. 65-70
[K62J Koch, W.H., Thyristor controlled pulsating field reluctance motor system.
Electric Machines and Electromechanics 1977, pg. 201
[K63J Kõllensperger, D., Die Synchronmaschine ais selbstgesteuerter Strornrichter
motor. Siemens Zeitschrift 1967, pg. 830
[K64J Kõrber, J., Teich, W ., Lokomotiven mit Drehstromantriebstechnik zur IVA
79, Beispiele eines breiten Anwendungsgebietes. Elektrische Bahnen 1979,
pg. 151
[K65J Kolar, J.W., Ertl, H., Zach, F.C., Influence of the modulation method on the
conduction and switching losses of a PWM converter system. IEEE Trans.
Ind. Appl. 1991, pg. 1063
[K66J Konhauser, W ., Naunin, D., Digital control of a subsynchronous converter
cascade by microcomputers. Microelectronics in Power Electronics and Elec
trical Drives, Darmstadt 1982, ETG-Fachberichte 11, pg. 369
[K67J Koníshi, T ., Kamiyama, K., Míyahara, Y., Ohmae, T., Sugimoto, N., An
application technology of microprocessor to adjustable speed motor drives.
IEEE lnd. Appl., Ann . Mect. COIlf. ltec. 1978, pg. 669
(1<(iR] 1<(J 1l i:-lh i , 'I'., ICaTlliyama, K. , OIIlIl ;W, '1' " A pcrfonnancc analysis of micro I)l'o(',',;:;or hw ;cd cOlll.rol HY S I."IlI :; :q '1,1 ind 1,0 iLdj II~LlLhl,' spepd motor uri ves. '11'; 1':1': 'l'1>' UI H. 11111. A PI'I. I!lHO , PI'" TiH
I" (;01
I'Í ""1'''' 11 '' II, 1\ ., M("' I'~I II"'(}II, II , I II ,'ul.l'lI,II":I" :1 A "n,"i!"1I von WJlI 'I'I\III. d""...h
1(llk,'""cI" Si'I ... , "II SI "",,'" 11 i',.. It ,1( l"il'l. 11111.' . 1'1', ·.~I H
Bibliography
Bibliography
~:1O
431
11<70]
[K93] Kume, T., Sawa, T., Yoshida, T., Sawamura, M., Sakamoto, M., A high
102 11\72] Koseki, T., Morizane, T., Ohsaki, R., Masada, E ., Novel linear induction drive: control scheme and converters. Proc. EPE 91, Firenze, pg. 1-481
[K94] [K95]
Korb, F., Die Gefáf3folgesteuerung, ein Mittel zur VerminderW1g der Blindlei
stung von Stromrichteranlagen. ETZ-A, 1958, pg. 273
11\71] Korb, F., Drehzahlgeregelter Positionierantrieb. BBC Nachrichten 1968, pg.
1 1\
7:1] Koyama, M., Yano, M., Two degrees of freedom speed controller using ref
11<71\]
11\7G]
(76]
11
11<77] 1'<78] 11< 7!)] 11(1'0] '" H !)
I" H'lJ II-:iUI
erence system model for motor drives. Proc. EPE 91, Firenze, pg. 3-60 Krafka, P., Bünte, A., Grotstollen, R ., Comparison of induction machine control with orientation on rotor flux or stator flux in a very wide field weakening region. Proc. EPE 95, Sevilla, pg. 3.486-3.491 Kranenburg, R., Klaassen, R., Torri, G., Gatti, E., Righ performance induc tion machine drive for a fiying shear, Proc. EPE 95, Sevilla, pg. 3-481 Krause, J ., Zur Regelgüte von Thyristorstromrichter-Umkehrantrieben. Messen, Steuern, Regeln 1969, pg. 430 Krause, P.C., Method of multiple reference frames applied to the analysis of symmetrical machinery. IEEE-PAS, 1968, pg. 218 Krishnan, R., Stefanovic, V.R., Control principies in current source induction motor drives. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1980, pg. 605 Krishnan, R., Control and operation of PM synchronous motor drives in the fi.eld- weakening region. Proc. IECON 93, Maui, pg. 745 Krüger, L., Naunin, D., Dynamic modelling and model based control of an induction machine. Proc. EPE 97 Trondheim, pg. 3.819-3.824 Krüger, M., Bordstromversorgung mit schnellaufendem Synchrongenerator, Diss. TU Braunschweig, 1992 Krzeminski, Z., Speed and rotor resistance estimation in observer system of induction motor. Proc. EPE 91, Firenze, pg. 3-538 Kubo, K ., Watanabae, M., Ohmae, 1., Kamiyama, K., A software-based specd regulator for motor drives. Proc. Int. Power Electr. Conf., Tokyo 1983, p~.
ill! "lIlinta, R ., Speed sensorless field-oriented control of induction motor with roLor resistance adaptation. IEEE Trans. Ind. Appl. 1994, pg. 1219 11\ K(l[ I\iibler, E., Der Strornrichtermotor. ETZ-A, 1958, pg. 15 I" H'i I I
I\IIIIIP, '1'., lwalcllne, T., Sawa, T., Yoshida, T., Nagai, 1., A wide constant power filu ge v.,<:I.or-controllccl AC motor drive using winding challgeuvt:r l(~chlliq\U ' . II ~E E -I'rall~.
1'( H:l 1
[LI]
[L2]
[L3J
[L4]
[L5J [L6J [L7] [L8]
1500
II, H'1f (, IIbota, R., Matsuse, K., Disturbance torque estimation for field oriented
I" !)l1
[K96]
Ind. Appl. 1991, pg. 9:H '1'" Snllo
[L9] [L10] [L11]
[L12]
speed vector controlled spindle motor drive with c10sed transition between with and without enGoder control. IEEE Trans. Ind. Appl. 1992, pg. 421 Kunz, U., Eine digitale zeitoptimale Lageregelung. Diss. TR Darmstadt 1971 Kurz, R., Hochgenaue Drehzahlregelung mit Rilfe des Phase locked-Ioop. Elektrie 1978, pg. 658 Kusko, A., Application of microprocessors to AC and DC electric motor drive systems. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1977, pg. 1079 Labahn, D., UntersucbW1gen an einem Stromrichtermotor in sechs-und zwolfpulsiger Schaltung. Diss. TU Braunscbweig, 1961 Lauger, A., The commutatorless DC motor with three-phase current excita tion. 2 nd IFAC Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 619 Lagasse, J., Prajoux, R. Behaviour of control systems including controlled converters, especially rectifiers: A review of existing theories, survey. 1st IFAC Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1974, pg. 1 Laithwaite, E.R" Kuznetsov, S.B., Development of an induction machine commutated thyristor inverter for traction drives. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1980, pg. 580 Lampersberger, M., Weiss, R., Ebner, A., Scbmidhofer, A., Improving the performance of a rotating tie-frequency converter control. Proc. EPE 99, Lausanne Langer, J., Umrichterspeisung von Synchronmotoren fur Rohrmühlen. BBC Mitteilungen 1970, pg. 112 Langweiler, F., Richter, M., Flu13erfassung in Asynchronmaschinen. Siemens Zeitschrift 1971, pg. 768 Larabi, A., Grenier, D., Labrique, F., A simplified vector control scheme for an induction motor revisited. Proc. Workshop on Advanced Motion Control (AMC) 98, Coimbra, pg. 24-29 Lataire, P., Induction motor drives with power transistor inverters. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1978, pg. 763 Lataire, Ph., Maggetto, G., Modern teaching on variable speed induction motor drives. Proc. EPE 89, Aachen, pg. 1445 Lataire, Ph., Van Muylem, R., Maggetto, G., Experiments for teaching on current- source inverter-fed induction motors with indirect flux control. Proc. EPE 91, Firenze, pg. 3-612 Latzel, W ., Berechnung der Laststorung von drehzahlgeregelten Gleichstro mantrieben 'mit geschachteltem Stromregelkreis. Regelungstechnik 1965, pg. 375
[L13] Lauffer, H., Die Drehstrommaschine mit polradwinkelabhangig eingepragten Lauferstromen. AfE 1968, pg. 289
[L14] Lawrenson, P.J., Ray, W.F., Davis, RM., Stephenson, J.M., Fulton, N.N., Blake, R.J., Controlled-speed switched-reluctance motors: present status and future potential. DrivesjMotorsjControls 82, Leeds [L15] Lecocq, D., Lataire, Ph., Study of a variable speed, double fed induction motor drive system with both stator- and rotor- voltages controllable. Proc. EPI<; 91, Firenze, pg. 2-337 [LlG] L,·", D.e., SIII, S.K., Park, M.n., IIigh performance current regulator for a li.. I
1;{2
Bibliography
Bibliography
Le-Huy, H., Perret, R, Roye, D., Microprocessor control of a current fed synchronous motor drive. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1979, pg. 873
IL l8] Leídhold, R., Garcia, G., Losses minimízation in a variable speed field oriented controlled induction motor. Proc. IECON 98, Aachen, pg. 865-870
II, 19] Leitgeb, W., Zur Bemessung drehzahlveriinderlícher Antriebe konstanter Leistung mit stromrichtergespeisten Drehfeldmaschinen. ETZ 1973, pg. 584
II :20] Leímgruber, J., Stationary and dynamic behaviour of a speed controlled nd
synchronous motor with cos li or commutatíon limít line control. 2 IFAC
Symp., Control ín Power Electronícs and Electrícal Drives, Düsseldorf 1977,
pg. 463
11,21] Leonard, H.W., A new system of electríc propulsíon. Trans. AIEE 1892, pg. 566
11.22] Leonard, H.W., Volts vs Ohms - Speed regulatíon of electríc motors. Trans. AIEE 1896, pg. 377
IL23] Leonhard, A., Asynchroner und synchroner Betríeb der der allgemeinen dop-
peltgespeísten Drehstrommaschíne. AfE 1936, pg. 483-502
IL21J Leonhard, A., Kennlinien von Gleichstrommotoren. Elektrotechník und Maschínenbau 1942, pg. 261
IL 2 ~,] Leonhard, A., Anlassen von Asynchronmotoren über KUSA-Wíderstand . Elektrotechník und Maschínenbau 1943, pg. 122
11 ,2(i] Leonhard, A. , Die Asynchronmaschíne bei Lastiinderungen. Elektrotechnik und Maschínenbau 1965, pg. 3
11,27] Leonhard, A., Die Drehstrommaschine mit eingepriigten Sekundiirstromen.
ETZ-A , 1968, pg. 97
ILn] Leonhard, R ., Schrõder , D., New precalculating current control strategy for
DC drives. Proc. EPE 87, Grenoble, pg. 659
IL2!)\ Leonhard, W., Elements of reactor controlled reversible índuction motor
drivcs. AIEE Trans. 1959, Pt II, pg. 106
11,:\1)1 Ll,lHlhard, W., Digitale Regelung und Programmregelung ín Digitale Sig
lIalvcrarbeitung in der Regelungstechník. VDE-Buchreíhe, Vol. 8, 1962, pg.
ILt7]
101
LeollllUrd, W., Müller, H ., Ein stetig wirkender digitaler Drehzahlregler. I':'I'Z- A, 1962, S.381
11,:\:21 Lco llh ard, W ., Ziihlende Rechenschaltungen für Regelaufgaben. AfE 1964,
pt; . 215
I' ,:\:\\ Lcouh ard , W., Regelkreise mit symmetrischer Übertragungsfunktion. RegeIUIlgstechnik 1965, pg . 4
11.:\'1] I,conhard, W ., Dynamic behaviour of a nonlinear sampled-datardcontrol
luop using pulse width controlled thyristors as power amplifiers. 3 IFAC Con grcss , London 1965, Paper 34 C.
11.:lf, I Lcollhard , W., Regelkreis mit gesteuertem Stromrichter ais nichtlineares Ab l.a"~ Lproblem . ETZ 1965, pg. 513
11,:\(i! Lcollhard, W., R egelprobleme bei der stromrichtergespeisten Drehstrom ASy llchrollmaschine. VDE-Fachberichte 1968, pg. 50
I' ,:\71 I ,(,ollhard , W., Cascade control, a versatile structural principie for t echnieiLI
a lld cyl!( ,rIlcLical conl.rol I;'ysterns. COlltrol iJml Inst.ruments Syst.erns 1970,
I' ,:\II
[L40] Leonhard, W., Field orientated control of a variable speed alternator con
nected to the constant frequency line. IEEE Conf.Rec. Control in Power
Systerns, College Station 1979
[L41] Leonhard, W., Elektrische Regelantriebe für den Maschinenbau. VDI-Zeit
schrift 1981, pg. 419
[L42] Leonhard, W., Microcomputer Control of High Dynamic Performance AC
Drives - A Survey, Automatica 1986, pg. 1
[L43] Leonhard, W., Adjustable speed AC drives. Proc. IEEE 1988, pg. 455
[L44] Leonhard,W., 30 years space vectors, 20 years field orientation, 10 years
digital signal processing with controlled AC drives. EPE-Journal 1991, pg.
13, pg. 89
[L45] Leonhard, W ., Field orientation with controlled AC drives - the natural
solution to a travelling wave problem. Automatika 1995, Zagreb, pg. 5-9
[L46] Leonhard, W ., Electrícal drives - universal sources of mechanical energy.
Proc. Workshop on Advanced Motion Control (AMC) 96, Tsu City, pg. 7-14
[L47] Le Pioufle, B., Louis, J .P., Influence ofthe dynamícs ofthe mechanical speed
of a synchronous servomotor on its t orque regulation. Proc. EPE 91 , Firenze,
pg. 3-412
[L48] Lessmeier, R., Schumacher, W., Leonhard, W. , Microprocessor-controlled
AC servo drives with synchronous or induction motors: Which is preferable?
IEEE Trans . Ind. Appl. 1986, pg. 812
[L49] Lessmeier, R., Digitale Signalverarbeitung für mehrachsige Antriebssysteme
mit Drehstrom-Stellmotoren. Diss. TU Braunschweíg 1989
[L50] Letas, H.H., Leonhard, W ., Dual axis servo drive in cylindrical coordinates
using permanent magnet synchronous motors with microprocessor control.
Conumel 83, Toulouse 1983, pg. II-23
[L51] Letas, H.-H., Míkrorechner-geregelter Synchron-Stellantrieb. Diss . TU
Braunschweíg 1985
[L52] Leví, E., Vuckovíc, V., Effects of maín flux saturatíon on behavíor of a stator
and airgap flux oríented induction machine. Proc. EPE 91, Firenze, pg. 3-618
[L53] Leví , E., Adaptíve decoupling circuit for universal field oriented controller
operating ín the airgap flux frame. Proc. EPE 93, Brighton, pg. 5-57
[L54) Leví, E., Vuckovic, V., Sokola, M., Rotor flux estimation in vector controlled
índuction machínes incorporating the iron loss compensatíon. Proc. EPE 95,
Sevilla, pg. 3.997-3.1002
[L55] Levi, E., Sokola, M., A novel saturation adaptive rotor flux estimator for ro tor flux orientated induction machines. Proc. EPE 97 Trondheim, pg. 1.518 1.523 [L56] Li, Y ., Jiang, L., Sensorless control of PMSM with an adaptive observer. Proc. EPE 99, Lausanne [L57] Li, Y .D ., de Fornel , B., Pietrzak-David, M., New approach for flux control of
PWM fed induction motor drive. Proc. IEE-Power Electronícs and Variable
Speed Dríves, London 1988 , pg. 301
[L58] Liao, F., Sheng, J. , Lípo, T.A ., A new energy recovery scheme for doubly
fed , adjustable-speed índuction motor dríves . IEEE Trans. Ind. Appl. 1991,
pg.728
[1.59J Li enall, W ., CommuLation modes of a current. source inverter. 2 nd IFAC Sylll[). , COJlt.rol iII Power Eledrollics alld I': kctrical Drives, Düsseldorf ]977 ,
pr( . :2K
II,:\HI I,, 'ollh(ll"d , W., ()[lt.illlak ZiPlHLC\lcrull i!, cilll ; ~:i k rii.ft.<:frei IH' weglic.hcll Mnssc[I ,:Iii ~
pUIII
1.", ,,01' 1" '.1, W " Z,'il.opl.iIIl ILi, ' Sc·!tc 'r"llrc"I'.C ·)'"If', . I\ "I~ 11)'70, PI~ ·
(j
I
433
[),(;01
p t; . :.! I!) )'i " rHtII , W. , Miill or- J[dIIll JUIIl, 1\ ., SllIlCl <1ll1y, )\. e it ., P()Wt ~r (", ()lIv""t.c:rH ro r . r,· c ·. l h'l~ Jl/l.Y JIC 'h ru lI""'i I. rll(' l.i,," IIII,j,' ''·H o\' 11;''1 \1, · ph JIII" I\C v.,h i dc· ~ 1. I'rne . 1111. . ~ "" d ' ·'''ICI" c'I., ,,· I '"wc ' r ( ~OIlV '· It., \ r (:",, 1'. I(, 'c ' 111'/ 7, 1' 1\. J~ IH)
~:'1
Bibliography
Bibliography
11.61] Lin, A.K., Koepsel, W.W., A microprocessor speed control system. IEEE
Trans. Ind. Electr. and Control Instrumentation 1977, pg. 241
11.62] Lin, F.J., Liaw, C.M ., Control of indirect field-oriented induction motor
drives considering the effects of dead-time and parameter variations. IEEE
Trans. Ind. Electr. 1993, pg.486
II,li:{] Lipo, T .A., Krause, P.C ., Stability analysis of a rectifier inverter induction
motor drive. IEEE Trans. Power Appl. Syst. 1969, pg. 55
IUi1] Lipo, T.A ., State variable steady state analysis of a controlled current in
duction motor drive. IEEE Trans. Ind. Appl. 1975, pg. 704
II,iiS] Lipo, T.A., Analysis and control of torque pulsations in current-fed induction
motor drives. Proc. Power Electronics Specialists Conf. 1978
11.1;(i] Lipphardt, G., Ruth, S., A new field supply concept for industrial adjustable
speed DC drives. Proc. EPE 99, Lausanne
II ,fi7] Liu, C ., Li, C., Wei, L. Li, F., Improvement of speed sensorless control
method of induction motor. Proc. IECON 98, Aachen, pg. 986-990
ILliS] Liu, S., Stiebler, M., A continuous discrete time state estimator for a syn
chronous motor fed by a PWM voltage source inverter. Proc. EPE 89,
Aachen, pg. 865
[1.69] Li, Y., Lee, F.C ., A comparative study of zero-current-transition (ZCT)
schemes for three-phase inverter applications. Proc. CPES Seminar 1999,
Blacksburg, pg. 143-151
IL 70] Lockwood, M., Fox, A.M., A novel high power transistor inverter. Proc. Int.
Power Electronics Conference, Tokyo 1983, pg. 637
IL7 l] Lbser, F., Sattler, Ph.K., Optimizing of the effi.ciency by the control of
inverter-fed induction machines especially regarding saturation and heat ef
fccts. 3 rd IFAC Symp. Control in Power Electronics and Electrical Drives,
Lausanne 1983, pg. 25
!"n] Lorenz, R.D. , Van Patten, K.W ., Righ resolution velocity estimation for
ali-digitai, AC servo drives . IEEE Trans. Ind. Appl. 1991, pg. 701
11,7:11 LOf CIl Z, R .D., Yang, S.M., Effi.ciency-optimised flux trajectories for closed
cycle operation of field-orientation induction machine drives. IEEE Trans.
111<1. i\.ppl. 1992, pg. 574
\I ,'1'1\ Lorun, L., Stator parameters influence on the field-oriented control tuning.
I'roc. EPE 93 , Brighton, pg. 5-79
\1 :/:,1 1,(llIis, J.P., Nonlinear and linearised models for control systems including
sLa.Li c: converters. 3 rd IFAC Symp. Control in Power Electronics and Electri
,:a! Drives, Lausanne 1983, pg. 9
11.f1i1 Luw , K.S., Deng, Y.Z. , Guo, X.L., Two-degree-of-freedom control of a PMSM
drive without a mechanical sensor. Proc. IECON 98, Aachen, pg. 498-502
li :{'fl Low, T.S ., Lee, T .R ., Chang, K .T ., A nonlinear speed observer for perma lIent-magnet synchronous motors. IEEE Trans. Ind. Electr. 1993, pg.307 I1.'7 HI Luot;ke, 1., Arias, A., Jayne, M.G ., Improving direct torque control of induc Lion motors. Proc . EPE 99, Lausanne
II :7!1] Ludwig, E., German Patent 1.072.704, 1960
ILHOI Luenberger, D .G., Observing the state of a linear system. IEEE Trans . Mil.
l::k cLr. 1964 , pg. 74
IMII Mnnr" RIl., Characteristics and application of current source slip regulated
1\<: indlldioll motor drives . IEEE Ino . Cen. Appl. Ann. Meet. COllf. Ree . 1')"11, pg. H
IM :!I MllI'dll, /\ ., Jluai Chill, '1' ., 'l'l\lIaku, II. , 1<01(IL, '1'., OhLllni, '1' ., 'l'oda'y'H A<;
driv I1 C"r llldl wt.rl ld applicrLl.ioJl iu .J :",lILlI . I' ro,". 1';1 ' 1'; !II, Firl'Il'l.() , pf.\. 'J-(iIH
IM :\I
MI" "', .1 , M... llu ·I), ·,·II , .I ., /\d l\p l.iv n 111;)( ol ll "'rv" f I~,r I.IJ I
II. I V" II \V IU.
I ' IIII /II H' ! '" dYll lull i t'
Ii,·",,,.rl'·:I" illdllcl.ioll
l' "r l~ IIIII I" III ' I ~ . l'I'tH' , 1':1'1 ';
UUI
III"
I jll.lHlI lIlIlC ~
435
[M4] Magyar, P., The dynamic behaviour of the current control loop of a
microcomputer-controlled DC drive in discontinuous current mode opera tion. Microelectronics in Power Electronics and Electrical Drives, Darmstadt
1982, ETG-Fachberichte 11, pg. 183
[M5] Magyar, P., Schnieder, E., Vollstedt, W., Digitale Regelung und Steue
rung einer stromrichtergespeisten Gleichstrommaschine mit Mikrorechner.
Regelungstechnik 1982, pg. 378
(M6] Maikath, R ., Putz, U., Saupe, R., An ali digital control for the LCI feeding
a synchronous machine. EPE 91, Firenze, pg. 4-586
[M7] Malesani, L., Tenti, P., Gaio, E., Piovan, R., Improved current control tech
nique of VSI PWM inverters with constant modulation frequency and ex
t ended voltage range. IEEE Trans. Ind. Appl. 1991, pg. 365
[M8] Malesani, L., Tomasin, P., PWM current control techniques of voltage source
inverters- a survey. Proc. IECON 93, Maui, pg. 670
[M9] Manninen, V., Application of direct torque control modulation technology
to line converter. Proc. EPE 95, Sevilla, pg. 1.292-1.296
[MIO] Martins, C.A., Carvalho, A.S., Roboam, X., Meynard, T.A., Evolution of
induction motor control methods and related voltage-source inverter topolo gies. Proc . Workshop on Advanced Motion Control (AMC) 98, Coimbra, pg.
15-20
[M11] Marwali, M.N., Keyhani, A., Tjanaka, W., An integrated digital signal pro cessor based system for induction motor vector control using off-line param eter identification. Proc. EPE 97 Trondheim, pg. 3.782-3.787 [M12] Mathys, S., Microprocessor-based multimode synchronous pulse width mod
ulation. Microelectronics in Power Electronics and Electrical Drives, Darm
stadt 1982, ETG-Fachberichte 11, pg. 237
[M13] Matsui, N., Sensorless operation of brushless DC motor. Proc. IECON 93 ,
Maui, pg. 739
[M14] Matuonto, M., Monti, A., Righ performance field oriented control for cycloconverter-fed synchronous machine: from theory to aplication. Proc. EPE 95, Sevilla, pg. 3.458-3.463 (M1~ Mauch, K ., !to, M.R., A multi-microprocessor AC drive controller. IEEE
Ind. Appl. Ann. Meet. Conf. Rec., 1980, pg. 634
[M1~ Maurer, F., Vergleich verschiedener Zündsteuerverfahren für netzgeführte
Stromrichter. ETZ-A, 1974, pg. 50
(M1~ Maurer, F ., Stromrichtergespeiste Synchronmaschine ais Vierquadrant-Re
gelantrieb. Diss. TU Braunschweig 1975
st [M1~ Mc Laren, S.G., Direct digital control of thyristors. 1 IFAC Symp., Control
in Power Electronics and Electrical Drives, Düsseldorf 1974, Vol. 2, pg. 147
[M1~ McMurray, W ., SCR inverter commutated by aTI auxiliary impulse. IEEE
Trans. Communications and Electronics, 1964, pg. 824-829
[M2~ McMurray, W., Thyristor corrunutation in DC choppers - a comparative
study. IEEE Trans. Ind. Appl. 1978, pg. 547-558
[M2~ Meiners, U., Beitrãge zur adaptiven Regelung elektrischer Antriebe. Diss.
TU Braunschweig, 1996
[M2~ Mertens, A., Harmonic distortion in three-phase inverters controlled by syn
chronous sigma-delta-modulation . Europ. Trans. on Electr. Power Engng.
1992, pg. 351
(M2:1] Ml'sl.ha, 1,.1< ., Evans, P. n., 0l'li"lisaL ion of losses in PWM inverters . Proc .
1I~ !':, P"w"r f>:k.dro"icH Iwd VILri nl.ll' S l' l~, : d Drives, London 1988, pg. 394
IM'.!>!I · M"'y"r, II ., V'·'·I'.I .. ich Jd ILqll i ~t ' h"r ,,,,,I lI11lde rlll :r H.~w~ lv erfahr e n am Beispi el
,·ill"'" ~;,·IiI\.,·"I.rt \ rd'·'· 1LIlIIlj('· . 111 1111. '1'1 1 IIrn,tll,"dlw,·iJl; I!J'J:'
,1:\(;
Bibliography
IM25) Meyer, H.G., Static frequency changers for starting synchronous machines. BBC Review 1964, pg. 526 IM26] Meyer, M., Mõltgen, G., Wesselak, F., Der Quecksilberdampf-Stromrichter ais Stellglied in Regelkreisen. Siemens Zeitschrift 1960, pg. 592 11\t127] Meyer, M., Über die untersynchrone Strornrichterkaskade. ETZ 1961, pg. 589 IM2H] Meyer, S., Sondermann, E ., Mehrmotorenantrieb für Rotationsdruckmaschi nen. Siemens Zeitschrift 1977, pg. 387 1M2!)] Michels, K., Fuzzy control for electrical drives ? Proc. EPE 97 Trondheim, pg. 1.102-1.105 IM:IO) Miki, 1., Nakao, O ., Nishiyama, S., A new simplified current control method for field- oriented induction motor drives. IEEE Trans. Ind. Appl. 1991, pg. 1081 IM:l1] Millet, C., Leroux, D., Li, Y ., Feuvrie, B., Bergmann, C., Modem experi mental tools for parameter identification and field-oriented control design of a low power induction machine. Proc. EPE 97, Trondheim, pg. 4.597-4.602 IM:\2] Miti, G.K., Renfrew, A.C., A microprocessor-based current profiling scheme for field-weakening applications in brushless DC motors. Proc. EPE 99, Lau sanne IM:n] Miyashita, 1., Ohmori, Y., Improvement of robustness on speed sensorIess
vector control of induction motor. Proc. EPE 91, Firenze, pg. 4-660
IM:\~] Miyashita, I., Ohmori, Y., A new speed observer for an induction motor
using the speed estimation technique. Proc. EPE 93, Brighton, pg. 349 IM:\:'] Mittenzwei, K., Probst, E., Thyristorgespeiste Fõrdermaschinen. BBC Nach richten 1976, pg. 153 IM:lliJ Mõltgen, G., Neuffer, 1., Stand der Technik der Antriebe mit netzgeführten SLromrichtem. ln Energieelektronik und geregelte elektrische Antriebe, V08-Buchreihe Vol. 11, 1960, pg. 239 11\11:\71 Müll.gcn, G., Salzmann, Th., Power factor and current harmonics with cyclo COlIv(~rl.crs fed by a three-phase supply. 2 nd IFAC Symp., Control in Power 1·;I,"("I.I"O"iC5 and Electrical Drives, Düsseldorf 1977, pg. 181 1rv1:l kl MIII',hl'f'lli, n., Adams, G.E., Hoft, R.G., Performance of alO HP switched ,""I"dlLllce motor and comparison with induction motors. IEEE Trans. Ind. fi ppl. 1991, pg. 531 IM: I! JI rvl,"'rschell, J., Signal processor based field oriented vector control for an illdllcLion motor drive. Proc. EPE 91, Firenze, pg. 2-145 IM 'IIII MocrschelI, J., Tursini, M., A new vector control scheme for inverter-fed 1'("f1f1cnent magnet synchronous motor using DSP. Proc. EPE 91, Firenze, pg. 1\ -683
IM'III Mnk, n.s., Kim, J.S., Kim, R.Y., Park, M.H., Sul, S.K., A stator fiux ori ""taLion speed control of induction motor without speed sensor. Proc . EPE !) I, Fircnze, pg. 4-678 11\'1.121 Mukrytzki, B., Pulse width modulated inverters for AC motor drives. IEEE ','rall s. lud. Gen. Appl. 1967, pg. 493 IM 'I:II Mnkryl.zki, H., The controlled slip static inverter drive, IEEE Trans. lnd. ÂI'PI. 1!J6H, pg. 312 IM -I,q Mo"s,,, M., MüJler,V., Cerv, H., Schulze, S., Modelling and dynamic simuln I.illll III' ckctrOlllechanical systems of equipment in the rolling rnill field. Prur. . Wor!,); lIo!, 011 Advallcvd Motion Control (AMe) !Jl) , CoiJllbra, pg. I\(H'--I\'/,j IM -Ir,1 Mo""il'lL, J.C., I,ipo , T .A., nl ;L~k", V., Sill'!"" ..Ili (:i(:IICY lIIaxilllisl"r ror 'UI ,d,i " "I.ld,I,· r""qll("II:Y ill
Bibliography
437
[M46) Moreira, J .C., Lipo, T.A., Modelling of saturated AC machines including air gap fiux harmonic components . IEEE Trans. Ind. Appl. 1992, pg. 343 [M47] Moreira, J.C., Hung, K.T., Lipo, T.A., Lorenz, R.D., A simple and robust adapti ve controller for detuning correction in field-oriented induction ma chines. IEEE Trans. Ind. Appl. 1992, pg. 1359 [M48) Moreno-Eguilaz, J.M., Peracaula, J., Real-time power measuring and moni toring for an efficient vector-controlled induction motor drive. Proc. IECON 98, Aachen, pg. 871-876 [M49] Morimoto, S., Sanada, M., Takeda, Y., Efficiency improvement of permanent magnet synchronous motor in constant power speed region. Proc. EPE 99, Lausanne [M50] Moschetti, A., Microprocessor-based PWM-system for inverter-fed squirrel cage motor traction drives. Microelectronics in Power Electronics and Elec trical Drives, Darmstadt 1982, ETG-Fachberichte 11, pg. 425 [M51] Moynihan, J., Kavanagh, R.C., Egan, M.G., Murphy, J.M.D., Indirect phase current detection for field oriented control of a permanent magnet syn chronous motor drive. Proc. EPE 91, Firenze, pg. 3-641 [M52) Mozdzer, A., Bose, B.K., Three phase AC power control using power tran sistors, IEEE Trans. Ind. Gen. Appl. 1976, pg. 499 [M53) MüIler, M., Roboter mit Tastsinn, Diss. TU Braunschweig, Vieweg Verlag Braunschweig- Wiesbaden, 1993 [M54) Mühlõcker, H., Elektrische Ausrüstung einer grofien Windkraftanlage. Sie mens-Energietechnik 1979, pg. 443 [M55) Murakami, T ., Yu, F., Ohnishi, K ., Torque sensorIess control in multidegree of- freedom manipulator. IEEE Trans. Ind. Electr.1993, pg. 259 [M56] Murphy, J.M.D ., Hoft, R.G., Howard, L.S., Controlled slip operation of an induction motor with optimum waveforms. 2nd Int. Conf. on Electr. Variable Speed Drives, IEE London 1979, pg. 157 [M57) Murphy, J.M.D., Egan, M.D., An analysis of induction motor performance with optimum PWM waveforms. Proc. Int. Conf. Electrical Machines, 1980, pg. 642 [M58) Mutoh, N., Sakai, K., Ueda, A., Tomita, H., Nandoh, K., Stabilizing methods at high frequencies for an induction motor driven by a PWM inverter. Proc. EPE 91, Firenze, pg. 2-352 [M59] Mutschler, P., Schulze, H., Control of current source inverter by single board computer for squirrel-cage motor drive of urban transport system. Microelec tronics in Power Electronics and Electrical Drives, Darmstadt 1982, pg. 417 [N1) Nabae, A., Otsuka, K., Uchino, H., Kurosawa, R., An approach to fiux con trol of induction motors operated with variable frequency power supply. IEEE Trans. Ind. Appl. 1980, pg. 342 [N2] Nabae, A., Akagi, H., Osanai, T., Unity power factor commutator-Iess motors based on a new commutation principie. Proc. Int. Semiconductor Power Converter Conf. 1982, pg. 342 [N3] Nagase, H., Matsuda, Y., Ohnishi, K., Ninomiya, H., Koike, T., High per formance induction motor drive system using a PWM inverter. IEEE-IAS Ann. Meeting, 1983 [N4] Nakano, T., Oksawa, H., Endoh, K., A high performance cycloconverter-fed synchronous machine drive system . Proc. Int. Semiconductor Power Con verter Conf. 1982, pg. 334 [NS] Nashild, M., note, Y., TTirçh IH~ l'rnrllllLll('" cllrn:llt.-contrulled l'WM tn~nsisLor ·i"vI'r!.' ·I· r"d IorllshksHS('rVo/llol.Clr . l' r()(' . 1111.. S,·/lli.,()/ldlll:l.or I'owc:r (;o/lv" rl,cr ( ;,,/11'. IU /l',!, PI\ ,
:\~O
,1:tH
Bibliography
Bibliography
Naunin, D., Ein Beitrag zum dynamischen Verhalten der frequenzges teuerten Asynchronmaschine. Diss. TU Berlin 1968 IN'71 Naunin, D ., The calculation of the dynamic behaviour of electric machines by space-phasors. Electric Machines and Electromechanics, 1979, pg. 33 IN HI Neft, C.L., Schauder, C.D., Theory and design of a 30 HP matrix converter. IEEE Trans. Ind. Appl. 1992, pg. 546 IN!JI Neuffer, 1., Stromrichterantriebe mit Momentumkehr. Siemens Zeitschrift 1965, pg. 1079 lN 1°1 Nguyen, M.T., Tu Xuan, M., Simond, J.J., Digital transient torque or force measurement for rotating or linear AC machines. Proc. EPE 95, Sevilla, pg. 1.319-1.321 lN IIJ Nielsen, P.E., Thomsen, E.Chr., Nielsen, M.T., Digital voltage vector control with adaptive parameter tuning. Proc. EPE 89, Aachen, pg. 313 lN 11J Nielsen, P.E., Mueller, K., Vammen, S., Speed and position AC servo drive based on current vector control. Proc. EPE 91, Firenze, pg. 3-635 INI:s] Nilsen, R ., Modeling, identification and control of an induction machine. Diss. NTH Trondheim, 1987 lN I~l Nishio, T., Minoguti, K., Uno, S., Hombu, M., Futami, M., Higuchi, M., Con trol characteristics of an adjustable speed generating system with a fiywheel excited by a DC link converter. Proc. EPE 97 Trondheim, pg. 2.695-2.700 II'J I:'] Nitschke, H.J., Der bürstenlose Motor, ein neuer universeller drehzahlregel barer Drehstromantrieb. AEG Mitteilungen 1973, pg. 73 lN HiJ Nobuyuki, K., Kamatani, M., Watanabe, H., Current source inverter drive Speed sensorless vector controlled induction motor. Proc. IECON 93, Maui, pg.983 lN 171 Nonaka, S., Neba Y., A PWM GTO current source converter-inverter system with sinusoidal inputs and outputs. IEEE Trans. Ind. Appl. 1989, pg. 76 lN IHI Nonaka, S., Neba Y., Quick regulation of sinusoidal output current in PWM eOllverter- inverter system. IEEE Trans. Ind. Appl. 1991, pg. 1055 lNl\ll Nonaka, S., Yamasaki, K., Hirohata, S., PWM current source converter illverter system for railways application. Proc. EPE 91, Firenze, pg. 4-72 IN ;~I)I NOJlaka, S., Kawaguchi, T., A new variable speed AC generator system using li hl'llshless self-excited type synchronous machine. IEEE Trans. Ind. Appl. I ~J!)2, pg. 490 I N ~ ii NOII,tka, S., Neba Y., Current regulated PWM-CSI induction motor drive sys l.em without a speed sensor. IEEE Trans. Ind. Appl. 1994, pg. 116 IN '2·.Q Nllllleroli, R., Gatti, E., Torri, G., Kranenburg, R., Four quadrant, large I'()wer, IGBT vector controlled adjustable speed drive. Design and test oProC. 1':1'1'; 95, Sevilla, pg. 1.508-1.512 IN',~: ti Nllfi , U., Huck, S., Systemdynamischer Zusarrunenhang zwischen Strom 'lIlldell, Spannungsmodell und Beobachter bei der Asynchronmaschine. etz Árchiv, 1990, pg. 227 I N/.'IJ Nystrolll, A., Hylander, J., Thorberg, K., Harmonic currents and torque I'lIlsatiollS with pulse width modulation methods in AC motor drives. Proc. 11·:I ~ .. Power Eleetronics and Variable Speed Drives, London 1988, pg. 378 I() II ()11I ~ rlllcicr, C., Kellermann, H., Brandenburg, G., Heinzl, J ., Sensorless 1i,·ld-Ol'iclIl.atcd specd control of a hybrid stepp er motor using an extended 1\ ,Lllll(l.ll lill.(')'. Proc. EPE 97 Trondheim, pg.1.238-1.2 /13 I( I,q (HII'I'IIII:i<:l', (:11., Realisierung CiHC:; posil.iOIlR-1l1l0 drch7.ll.hlgcregelten Scr vl);lIlI,ri,·,h,; III i tt."IH eint! Sch(~ il","liiHfnr-:-;chril,l.lII()to.r,.;. l'ro(. Sl':-J/I PC/
INli]
I)IUVI ~S \lH , Niiruh'TI~,
PI>. 171 - IHO
439
[03] Odai, M., Hori, Y., Speed control of a 2-inertia system with gear backlash us ing gear torque compensator. Proc. Workshop on Advanced Motion Control (AMC) 98, Coimbra, pg. 234-239 [04] Ogasawara, S., Akagi H., Nabae, A., The generalised theory of indirect vector control for AC machines. IEEE Trans. Ind. Appl. 1988, pg. 470 [05] Ogasawara, S., Akagi H., An approach to position sensorless drive for brush less DC motors. IEEE Trans. Ind. Appl. 1991, pg. 928 [06] Ogino, Y., Murakami, Y., Ohishi, K., Nakaoka, M., High performance ultra low speed control of inverter-fed linear induction motor using vector control scheme. Proc. EPE 93, Brighton, pg. 5-464 [07] Ohm, H.J., Oppermann, K., Raatz, E., Steuerung einer durchlaufenden Sage mit einem Kleinstrechner. AEG Mitteilungen 1976, pg. 266 [08] Ohmae, T., Matsuda, T., Suzuki, T., Azuwada, N., Kamiyama, K ., Kon ishi, T., A microprocessor-controlled fast-response speed regulator with dual mode current loop for DCM drives. IEEE Trans. Ind. Appl. 1980, pg. 388 [09] Ohnuki, T., Miyashita, O., Lataire, Ph., Maggetto , G., A three-phase PWM rectifier without volt age sensors. Proc. EPE 97 Trondheim, pg. 2.881-2.886 [010] Ohtani, T., Takada, N., Tanaka, K., Vector control of induction motor with out shaft encoder. IEEE Trans. Ind. Appl. 1992, pg. 157 [011] Ohyama, K., Asher, G.M., Sumner,M., Comparison of the practical per formance and operating limits of sensorless induction motor drives using a closed loop fiux observer and a full order flux observer. Proc. EPE 99, Lausanne [012] Oldenkamp, J.L., Peak, S.C., Selection and design of an inverter driven in duction motor for a traction drive system. GE-Report 83 CRD 191, 1983 [013] Olomski, J., Bahnplanung und Bahnführung von Industrierobotern. Diss. TU Braunschweig, Vieweg Verlag Braunschweig-Wiesbaden 1989 [014] Omori, Y., Nakanishi, T., Kobayashi, H., A speed sensorless spatial vector controlled inverter added an auto-measuring function. Proc. EPE 95, Sevilla, pg. 3.452-3.457 [015] Orlik, B., Zum Betriebsverhalten mikrorechner-geregelter und pulswechsel richter- gespeister Synchronmaschinen mit Permanenterregung. Diss. TU Braunschweig 1987 [016] Orlik, B., Economical inverter-fed AC drives with fully digital control by means of an 8 bit microcontroller. Proc. EPE 91 Firenze, pg. 4-578 [017] Orlik, B., Feldorientierte sensorlose Drehzahlregelung von Drehstrom-Asyn chronrnaschinen. Proc. SPS/IPC/DRIVES 94, Stuttgart, pg. 541-552 [018] Oumamar, A., Louis, J.P., EI-Hefnawy, A., Design of an optimal, auto adaptive current loop for DC motor. Realization with a hybrid device in cluding a microprocessor. 2 nd IFAC Symp ., Control in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 593 [019] Oyama, J., Mamo, M., Abe, T., Higuchi, T., Yamada, E., High frequency method of sensor elimination in interior permanent magnet motors. Proc. EPE 97, Trondheim, pg. 4.535-4.540 [PI] Pachter, M., Speed control of a field controlled d.c. traction motor. Auto matica 1981, pg. 627 [P2] Pagano, E., Perfetto, A., A comparison between operating conditions of inverter-fed asynchronous motors. 2 nd IFAC Symp., Control in Power Elec (,ronies (tJld Electrical Drives, Düsscldorf 1977, pg. 399 [1':1] l'aiC\~, )).Á ., Illduction nlOtor Hp(wd (;O lItrol hy ~ti1tor voltagc control. IEEE 'l'nuI", l'IlW('1' ÁI'I'1. S.YHI" I!)OH, pg. 6Wi jJ ,III
1'lIl1 l1 il.' '' '' 'II, II. ( !" NOrlllll, 1,.1-:., 1\111,111111 lilt"I' l lIlI" 'd tJ( 'II Hllrl,'~'~; 1:11111."01111' ali II'M 111"1,,,,' driv, '. 1'"",. 1':1'1'; 11 '1, '1''''11,1111'1111, 1'1', ·1 1,fi '{ ·I. IJ ~ ·
'110
Bibliography
Bibliography
\l'5] Papiernik, W., Teckentrup, M., Kartesische Bahnregelung rur Servoantriebe.
Proc. SPS/IPC/DRIVES 98, Nürnberg, pg. 353-362
Jl'tiJ Parasuram, M.K., Ramaswarni, B., Analysis and design of a current-fed in
verter. 2nd IFAC Symp., Control in Power Electronics and Electrical Drives,
Düsseldorf 1977, pg. 235
II '7] Patel, H.S., Hoft, R.G., Generalised techniques of harmonic elimination and
voltage control in thyristor inverters. IEEE Trans. Ind. Appl., Part I, Har
monic elimination 1973, pg. 310; Part II, Voltage control techniques 1974,
pg. 666
II 'H] Pedersen, J .K., Blaabjerg, F., Jensen, J .W., Thoegersen, P., An ideal PWM
VSI inverter with feedforward and feedback compensation. Proc. EPE 93,
Brighton, pg. 5-501
I!'!)] Peixoto, Z.M.A., Seixas, P.F., Parameter identification for induction ma chines at standstill. Proc. EPE 99, Lausanne 11'10] Pena, R .S., Clare, J.C., Asher, G.M., Implementation of vector control strategies for a variable speed doubly-fed induction machine for wind gener ation system. Proc. EPE 95, Sevilla, pg. 3.075-3.080 II' 11] Peng, F .Z., Fukao, T., Robust speed identification for speed-sensorless vector
control of induction motors. IEEE Trans. Ind. Appl. 1994, pg. 1234
li' I:2J Pereira, L.F.A., Halfner, J.F., Hemerly, E.M., Gründling, H.A ., A simulation
frarnework for flux estimation and vector control of induction machines.
Proc. IECON 98, Aachen, pg. 1587-1591
li' I:q Pereira, L.F.A., Haffner, J.F ., Hemerly, E.M., Gründling, H.A., Direct vec
tor control for a servopositioner using an alternative rotor flux estimation
algorithm. Proc. IECON 98, Aachen, pg. 1603-1608
II) I~] Peresada, S., Tilli, A., Tonielli, A., Indirect field-oriented control of induc
tion motor: New design leads to improved performance and efficiency. Proc.
IECON 98, Aachen, pg. 1609-1614
li' I ~'I I'eresada, S., Tilli, A., Tonielli, A., Robust active-reactive power control of
;t doubly-fed induction generator. Proc. IECON 98, Aachen, pg. 1621-1625
II' I(ii 1'".. il1, A.J., da Cunha, A.A., de Oliveira, M.A., Variable optimal PWM for
vollage source rectifiers and current source inverters. Proc. EPE 89, Aachen,
PI ;. 1279
II' I 'r l I', ~ rllg, S.S., Lai, Y .S., Liu, C.H ., Sensorless vector controller for induction
IllOl.or drives with parameter identification. Proc. IECON 98, Aachen, pg.
IOOS-1013
11'1 HI I'esse , G., Pagá, T., A permanent magnet synchronous motor flux control scheme without position sensor. Proc. EPE 97, Trondheim, pg. 4.553-4.560 II' I!)I I'cters, K.W., Theoretische und praktische Untersuchungen an einer reihen
schluBerregten, stromrichtergespeisten Reluktanzmaschine. Diss. TU Braun
schweig 1979
I!'/.IIJ P<"lers, W., SYNAX, ein Motion-Control-System für wellenlose Druck- und l'apierverarbeitungsmaschinen. Proc. SPS/IPC/DRIVES 97, Nürnberg, pg. :\77-389 IJ>:.! II ['el(, rsen, LR., Pulle, D.W.J ., Robust Kalman filtering in direct torque con
lrol. Proc. EPE 97, Trondheim, pg. 4.585-4.590
I!'n l I',·l,·.r>;ol1, B., Valis, J., Oscillations in inverter fed induction motor drivcs.
I'ro r.. EI'E 91 F'irenzc, pg. 2-369
II' ~!:\I I', ·l.n'H'II, II., 1nd llc\,ioll mil.chinc ::;pC(,([ flst.imuLioll; ol>sp.. vatioIl~ ou ohs"rvcr~.
I )iI'lH . '1'11 \.lIlId, I!)!JH
1I' ~· iI 1',·1.1'1"« :,,"k'l~ ", 1,.-(>', l'i('Lrz;alc- lliLVid, M., 1\"11 ÂII'II"", 1". , d" II"rll"l, II.,
11'I ',h 1" "'('''1'11'"11' ''' (',,"1.1',,1 r"r loil ~ h - p"W"r illd", l.i'II' 1I"... loi",' wil.l,,,"1. :11"'"''
[P25] [P26]
[P27] [P28] [P29] [P30] [P31]
[P32] [P33] [P34] [P35] [P36) [P37) [P38] [P39] [P40) [P41] [P42] [P43) [P44) [P1 f:i] [I'~(i]
1I',I'rl
441
sensor: choice and comparison of two methods. Proc. EPE 95, Sevilla, pg. 3.430-3.435 Pfalf, G., Zur Dynamik des Asynchronmotors bei Drehzahlsteuerung mit
verii.nderlicher Speisefrequenz. ETZ-A 1964, pg. 719
Pfaff, G., Time-response analysis of inverter-fed induction motors with con
trolled stator current. 2nd IFAC Symp., Control in Power Electronics and
Electrical Drives, Düsseldorf 1977, pg. 493
Pfaff, G., Scheurer, H.G ., Pararneterunempfindliche Regelung von Gleich
strommotoren im Feldschwii.chbereich. ETZ-A, 1969, pg. 400
Pfalf, G., Weschta, A., Wick, A., Design and experimental results of a brush
less AC servo motor. IEEE Ind. Appl. Ann. Meet. Conf. Rec. 1982, pg. 692
Pfaff, G., Wick, A., Direkte Stromregelung bei Drehstromantrieben mit Pul
swechselrichter. Regelungstechn. Praxis 1983, pg. 472
Pfaff, G., Segerer, H., Resistance corrected and time discrete calculation of
rotor flux in induction motors. Proc. EPE Aachen 89, pg. 499
Pfitscher, G.H., A microprocessor control scheme for naturally commutated
thyristor converter with variable frequency supply. 3 rd IFAC Symp. Control
in Power Electronics and Electrical Drives, Lausanne 1983, pg. 431
Phillips, K.P., Current source converter for AC motor drives. IEEE Trans.
Ind. Appl. 1972, pg. 679
Pienaar, A.L, A contribution to the solution of instability problems in supply
commutated power electronic converter-fed systems. Diss. Randse Africaanse
Universiteit Johannesburg 1980
Pietrzak-David, M., de Fornel, B., Universal PWM inverter fed asynchronous
drive with a dynarnic flux control. Proc. EPE 89, Aachen, pg. 1329
Plaetrick, H ., Digitale Lageregelungen. AEG Mitteilungen 1976, pg. 291
Plunkett, A.B., Direct flux and torque regulation in a PWM inverter
induction motor drive. IEEE Trans. Ind. Appl. 1977, pg. 139
Plunkett, A.B ., A current-controlled PWM transistor inverter drive. IEEE
Ind. Appl. Ann. Meet. Conf. Rec. 1979, pg. 785
Plunkett, A.B., Lipo, T.A., New methods of induction motor torque regula
tion. IEEE Trans. Ind. Appl. 1976, pg. 47
Plunkett, A.B., Turnbull, F.G., Load commutated inverter synchronous mo
tor drive without a shaft position sensor. IEEE Trans. Ind. Appl. 1979, pg.
63
Pohl, K., Rumold, G., Digitaler Regler in TTL-Technik rur genaue Drehzahl
und Drehzahlverhii.ltnisregelung. Siemens Zeitschrift 1972, pg. 335
Pollmann, A., Gabriel, R, Zündsteuerung eines Pulswechselrichters mit
Mikrorechner. Regelungstechnische Praxis 1980, pg. 145
Pollmann, A., A digital pulse-width modulator employing advanced modu
lation techniques. IEEE Trans. Ind. Appl. 1983, pg. 409
Pollmann, A., Comparison of PWM techniques. Microelectronics in Power
Electronics and Electrical Drives, Darmstadt 1982, ETG-Fachberichte, pg.
231
Pollmann, A., Ein Beitrag zur digitalen Pulsbreitenmodulation bei wechsel
richtergespeisten Asynchronmotoren. Diss. TU Braunschweig, 1984
Pollmann, A., Tosetto, A., Brea, P., Microprocessor based control system
ror rolling mills. Proc. EPE 91 Firenze, pg. 1-64
I'rajoll x, IL, Jalac.le, .T., Marpinard, J.C ., Mazankine, J ., A modelling method ror 1,11<' 1'('[lilviollr or (:OJlV(,r!.'TS opo:ral.illg in controlloops. 2nd IFAC Symp., (:olll.r,d iII I'"wn 1':I( ·(' t.rolli(" ,~ HllcI 1-;I,·cI.ri nd Ihiv,"s, Diisscldorf 1977 , pg. (i7 1'1<',111, 11 , 1',) Sl.i \ "'III I \~ 1I111 . , ·r drikhllll l\ d,ore -h ZlI fl l,lllld !: rt W' IIIIII (. 11.'·W·ltlllg
,: 1.. -.·111 011. I\lHII, 1'1'. 'J.'J.'l, :':1111
,1,12
Bibliography
11'18] Probst, U., Comparison of disturbance suppression strategies for servo
drives. Proc. EPE 93, Brighton, pg. 5-442
Ii'/f~)] Probst, E., Antriebstechnik in Konti- Walzwerken. BBC Nachrichten 1977,
pg. 485
1i'!;O] Profumo, F., Pastorelli, M., Ferraris, P., De Doncker, R .W., Comparison
of Universal Field Oriented (UFO) controJler in different reference frames.
Proc. EPE 91 Firenze, pg. 4-689
Ii'" IJ Putz, U., Thyristor industrial drives. Proc. Int . Semiconductor Power Con
verter Conf. Rec. 1977, pg. 340
II '!,:21 Pyrhõnen, J., Niemela, M., Kaukonen, J., Luukko, J ., Pyrhõnen, O., Tiiti nen, P., Vaananen, J ., Synchronous motor drives based on direct flux linkage contro!' Proc . EPE 97 Trondheim, pg.1.434-1.439 11'!dl Pyrhõnen, O., Analysis and control of excitation, field weakening and stabil
ity in direct torque controlJed electricaJly excited synchronous motor drives.
Diss. Lappeenrantta 1998
II'!;'11 Pyrhõnen, J., Pyrhõnen, O. , Niemela, M., Luukko, J., A direct torque con troJled synchronous motor drive concept for dynamicaJly demanding appli cations. Proc. EPE 99, Lausanne I') II Quang, Ng.Ph., SchneJle Drehmomenteinpragung in Drehstromstellantrie
ben. Diss. TU Dresden, 1991
Iq 0'4 1 Quang, Ng.Ph., Schõnfeld, R ., Dynamische Stromregelung zur Drehmo
rnenteinpragung in Drehstromantrieben mit Pulswechselrichter. Archiv für
Elektrotechnik, 1993, pg. 317
II ( II ltaatz, E., Der Einsatz von adaptiven Drehzahlreglern in der Antriebstech
lIik, AEG Mitteilungen 1970, pg. 375
III II n""tz, E., Der Einflul3 von elastischen Übertragungselementen auf die Dy
Il;unik geregelter Antriebe. AEG Mitteilungen 1973, pg. 205
II1 :q I(a,ducanu, D.C., Hochdynamisches Positionsregelsystem Axumerik . BBC
Nachrichten 1977, pg. 301
II (.'1 Iiapp , n., Examination of transient phenomena in induction machines,
(':\IIS<:
(·.."trol ~c heme. Europ. Trans. on Electr. Power Engng. 1993, pg. 397
11 1,1 HaHII'lIssen, H., Knudsen, M., Toennes, M., Inverter and motor model adap 1. lLti .. " aI, stand-stiJI using reference voltages and measured currents. Proc.
10;1'1': ~)!í, SeviJla, pg. 1.367-1.372
III iii 1(IIHI""ssen, K.S ., Thoegersen, P., Model based energy optimiser for vector
co"Lrolled induction motor drives. Proc. EPE 97 Trondheim, pg. 3.711-3.716
11('( 1 l(;lslIlu ssen, K.S ., Field magnitude adaptation in induction motors. Proc.
I': I' I': ~)7 'l'rondheim, pg. 3.922-3.926
IlU,! IL,s lllussen, H., Vadstrup, P., Boersting H., Nonlinear control of induction
IIlOl,orS: a performance study. Proc. Workshop on Advanced Motion Control
(AMe) !JS, Coimbra, pg . 109-116
I}( !II I( :' y, W.F., AI-Dahadly, LH ., SensorIess methods for determining the rotor
I'()sitioll of switched reluctance motors. Proc. EPE 93, Brighton, pg. 6-7
I" 10 1 li!:I"""", M., Modulare Echtzeitsimulation in der elektrischen Energietech
lIilL l'roc, EPE {<'irenze 91, pg. 1-74
II! I II IId,;I',.J .M ., Allard , B., Jorda, X., Perez, A ., Use of ASIC's in PWM tech
";ljll(,:') ror Jlower COllvcrtcrs. Proc. IECON 93, Maui, pg. 683
II! 1:' 1 11;1")('<:1<, C., (;"dakow f!ki, M., J)igitale MisdlUllGsregelung Systcrn DLEN,"
I)()MA'I', t\JI,;<: MitLI :il"llgl:n I~)(j:'\, PU. :': 17
1'11 :q 1('!"'lro , I.. A. S. , .J I,.co l,l-IlIL, ( :. L1 ., I' nFILIlH'kl' iLI,,1 :;(>1'( '<11 1 :~tillliLti()lI for illd" c
,,;"" '"!(f ·r,ill l 'll 1,/""·,, "u dY" H.llli c It)O"'''''. 1'1"11, ' 1':1'1,: !I !~ , !·;(·vill :L, I'I', I./fun ·
1
I.i,~ II
Bibliography
443
[R14] Ribickis, L.S., Ivbuls, U.V., Greivulis, J.P., Petrov, S.S., Static Kriimer drive
with specialIy designed voltage inverter. Proc. EPE 91 Firenze, pg. 2-301
[R15] Richardson, J. Bezanov, G ., A PWM 3-phase on-line controlIer for AC drives.
Proc. EPE Firenze 91, pg. 3-299
[R16] Richter, W., Microprocessor-controlIed inverter-fed synchronous motor
drive. Proc. IEE Conf. on Variable Speed Drives, London 1979, pg. 161
[R17] Richter, W., Einsatz eines Mikrorechners zur Regelung eines Stromrichter
motors bei Betrieb mit maximalem Leistungsfaktor. Diss. TU Braunschweig
1981
[R18] Richter, W., Microprocessor control of the inverter-fed synchronous motor.
Process Automation 1981, pg. 100
[R19] Riedo, P.-J ., Cascade digital control by state variable feedback method ap
plied to DC-motor. Microelectronics in Power Electronics and Electrical
Drives, Darmstadt 1982 ETG-Fachberichte 11, pg. 249
[R20] Riedo, P.-J., Cascade control by state variable feedback method applied to
a synchronous motor. 3 rd IFAC Symp. Control in Power Electronics and
Electrical Drives, Lausanne 1983, pg. 111
[R21] Riefenstahl, U., Einsatz von Mikrorechnern zur Steuerung und Regei ung
elektrischer Antriebssysteme. Elektrie 1978, pg. 245
[R22] Riefenstahl, U., Nguyen H., Bannack, A., Tensile force and angular synchro nism control for flexibly coupled drives by the example of a cage-stranding machine. Proc. EPE 99, Lausanne [R23] Riehlein , D., Getriebeloser Antrieb für eine Zementmahlanlage. Siemens
Zeitschrift 1971,S. 189
[R24] Riese, M. A microcontroJler implementation of speed sensorless field oriented
induction machine. Proc. EPE 97, Trondheim, pg. 4.476-4.479
[R25] Robert, P.Ph., Gautier, M., Bergmann, C., MiJlet, C., Feuvrie, B., Experi mentai identification of asynchronous machine. Proc. EPE 97, Trondheim, pg. 4.620-4.625 [R26] Robyns, B ., Buyse, H., Labrique, F., Comparison of the sensitivity of the
flux control to parameter uncertainties in two induction actuator indirect
field oriented control schemes. Proc. EPE 93, Brighton, pg. 5-402
[R27] Robyns, B., Buyse, H., Sensitivity of the flux control to parameter uncer tainties in induction motor direct F.O.C. strategies using a simple flux esti mation. Proc. EPE 95, SeviJIa, pg. 3.674-3.679 [R28] Robyns, B., Meuret, R., Sente, P., Current digital control influence on per formance of induction motor indirect field orientation control. Proc. EPE 99, Lausanne [R29] Rohner, R., Lineare Servomotoren im Maschinenbau. Proceedings SPS/IPC/
DRIVES 98, Nürnberg, pg. 533-539
[R30] Rojek, P., Bahnführung eines Industrieroboters mit Multiprozessorsystem.
Diss. TU Braunschweig, 1987
[R31] Rowan, T.M., Kerkman, R.J., Leggate, D., A simple on-line adaptation for
indirect field orientation of an induction machine. IEEE Trans. Ind. Appl.
1991, pg. 720
[R32] Rufer, A.-Ch., SchibJi, N., Briguet, Ch., A direct coupled 4-quadrant multi
levei converter for 16 2/3 Hz traction systems. Proc. Conf. Power electronics
and variable speed drives, IEE 1996, London, pg. 448-453
[R::!:\] Ilufr, M., Grotstollen, H., Identificat.ion of the saturated mutual inductance 01'
II
"r,,1' , 1':/'1': !)\ Bri ght.oll, 1'1( . ;,- 10:\ Sa l l"""., :';., I ,('111""1, ~;, ()V( 'n1Itoot
II"·tl,,,,1 1'," l"vVM
1,,\1, ·,1.('''1 1'1'" ,'
/1,11"
lilllil. ('yd"
1'1'1'1' CllI-r",,1. co"I.I·()1
1': 1'1': II I F lr" ,,'{.(·, PI'. . :\ 2 '1'{
444
[S2]
[S3]
[S4] [S5] [S6] [S7] [S8] [S9] [SI0] [S11] [SI2]
[SI3]
[SI4] [SI5]
[SI6] [SI7]
[SI8] [SI9] [S20] [S21]
[S22]
Bibliography
Bibliography
Saio, M., Tuusa, R., A vector controlled PWM current source inverter-fed induction motor drive with a novel rotor time constant estimation method. Proc. EPE 99, Lausanne Salvatore, L., Stasi, S., Tarchoni, L., A new EKF-based algorithm for flux estimation in induction machines. IEEE Trans. Ind. Electr. 1993, pg. 496 Salzmann, Th., Wokusch, R ., Direktumrichterantrieb für grofie Leistungen und hohe dynamische Anforderungen. Siemens Energietechnik 1980,pg. 409 Salzmann, Th., Kratz, G., Daubler, Chr., Righ power drive system with advanced power circuitry and improved digital control. IEEE Trans. Ind. Appl. 1993, pg.168 Santisteban, J .A., Stephan, R.M., Comparative study of vector control meth ods for induction machines. Proc. EPE 93, Brighton, pg. 5-390 Sathiakumar, S., Nguyen, M.T., Shrivastava, Y., Sensorless-speed control of induction motor. Proc. EPE 99, Lausanne Sato, N., A study of commutatorless motor. Electr. Engin . in Japan 1964, pg. 42 Sato, N., A brushless DC motor with armature induced voltage commuta tion. IEEE Trans. Ind. Gen. Appl. 1971, pg. 1485 Sato, N., Semenov, V.V., Adjustable speed drive with a brushless DC motor. IEEE Trans. Ind. Gen. Appl. 1971, pg. 539 Sato, N., Ishida, M., Sawaki, N., Some characteristics of induction motor driven by current source inverter. Proc. Int. Semiconductor Power Converter Conf. Rec. 1977, pg. 477 Sattler, P.K., Balsliemke, G., Experimental and theoretical investigations concerning the design of induction machines with hi~h power density with respect to the inverter supply and speed regulation. 2n IFAC Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 387 Sattler, Ph.K., Stiirker, K., Estimation of speed and pole position of an inverter fed permanent excited synchronous machine. Proc. EPE 89, Aachen, pg. 1207 Sawaki, N., Sato, N., Steady-state and stability analysis of induction motor driven by current source inverter. IEEE Trans. Ind. Gen. Appl. 1977, pg. 244 Schafer, U., Brandenburg, G., Compensation of Coulomb friction in indus triai elastic two-mass systems through model reference adaptive control. Proc. EPE 89, Aachen, pg. 1409 Schafer, U., Comparative study of inverter-fed drives with different motor types. Proc. EPE 95, Sevilla, pg. 1.502-1.507 Schauder, C., Adaptive speed identification for vector control of induction motors without rotational transducers. IEEE Trans. Ind. Appl. 1992, pg. 1054 Schierling, R., Self-commisioning, a novel feature of inverter-fed motor drives. Proc. IEE-Power Electronics and Variable Speed Drives , London 1988, pg. 287 Schlabach, L.A ., Analysis of discontinuous current in a 12-pulse thyristor DC motor drive. IEEE Trans. Ind. Appl. 1991, pg. 1048 Schnieder, E., Control of DC-drives by microprocessors. 2nd IFAC Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1977, pg. 603 Schnieder, E., Digitale Nachbildung stromrichtergesp eister Maschinell un ter b esonderer Berücksichtigung der wechselrichtergespeisten reilwIlschluOcr regLen Reluktanzrnaschine. Diss. TU Braull'i<"hw("ig I !)78 Sdlllicder, )'; ., Modl'llill f!; anel silllulllt.ioJ\ of" .ll1I11I1·ri,.n.II'y cOlIl.r"l k d t,hyriHI.or r,·d . driVl""' , 1' 1'(1(:. CONlfMI';I.I:lf) , p. 1I1/1 ~ - ~ : \
44G
~2~ Schonfeld, R., Einflufi der Getriebelose auf Stabilitat und Genauigkeit VO II Lageregelkreisen. Messen, Steuern, Regeln 1965, pg. 40 ~~ Schonfeld, R., Krug, R ., Niiherungsverfahren zur dynamischen BeschreibulI{-( der Strornrichterstellglieder. Messen, Steuern, Regeln 1972, pg. 246 ~2~ Schonfeld, R, Das dynamische Verhalten des Stromrichterstellgliedes im Lückbereich. Messen, Steuern, Regeln 1977, pg. 79 ~2~ Schonfeld, R., Entwicklungstendenzen der elektrischen Antriebstechnill. Elektrie 1981, pg. 451 ~~ Schonfeld, R ., Franke, M., Rasan, R., Müller, F., Intelligent drives in systew 'i with decentralised intelligence. Proc. EPE 93, Brighton, pg. 5-489 ~2~ Schonfeld, R., Control of nonlinear motion processes in processing machincs by individual electric drives. Proc. Workshop on Advanced Motion ConLrul (AMC) 96, Tsu City, pg. 447-451 [S29] Schonung, A., Stemmler, R., Geregelter Drehstrom-Umkehrantrieb mit gc:< teuertem Umrichter nach dem Unterschwingungsverfahren. BBC Nl1ch · richten 1964, pg. 555 [S30] Schonung, A., Stemmler, R., Static frequency changers with subhal"lllOJli ,. control in conjunction with reversible variable speed drives. BBC n .(:v i, ·w 1964, pg. 555 [S31] Schrader, A., Eine neue Schaltung zur Kreisstromregelung in Strornri,.ld.,·, anlagen. ETZ-A , 1969, pg. 331 [S32] Schroder, D. , Untersuchung der dynamischen Eigenschaften V O Il SI.I·(>Jt\ richterstellgliedern mit natürlicher Kommutierung. Diss. TR Dnl'lld. "dl. 1969 [S33] Schroder, D., Die dynamischen Eigenschaften von StromrichtersLcllgli .. d"" mit natürlicher Kommutierung. ETZ-A, 1970, pg. 242 [S34] Schroder, D ., Grenzen der Regeldynamik von Regelkreisen miL Sl.rolllril"l,I.'·1 stellgliedern. Regelungstechnik 1973, pg . 322 [S35] Schroder, D., Einsatz adaptiver Regelverfahren bei Regelkreise.ll IIIil. ~; I.I,,'" richterstellgliedern. VDE-Aussprachetag Freiburg 1973, pg. 81 [S36] Schroder, D., Selbstgeführter Stromrichter mit PhasenfolgcliiHcl III' '1". ,",01 eingepragtem Strom. ETZ-A, 1975, pg. 520 [S37] Schroedl, M., Sensorless control of permanent magnet synchronnll" ,".',,",11'''':. at arbitrary operating points using a modified "inform" flux m o do!. 1':111"1 '1' Trans. on Electr. Power Engng. 1993, pg. 277 [S38] Schroedl, M., Rennerbichler, D., Wolbank, T .M., Induction rno!.o,· dl'v, ' I", electric vehicles without speed and position sensors. Proc. EI'E 0:1, 111·il\l,I,,,,,. pg . 5-271 [S39] Schülting, L., Skudelny, R.-Ch., A control method for pr.rJu'tllt"liI. ''' ''1'.11 ''1, synchronous motors with trapezoidal electromotive for ce. J 'rlle. 1':1' I'; \11 . Firenze, pg. 4-117 [S40] Schütze, T ., Stronisch, V., Low floor trams with IGBT- ;1" l"v,·1 i " " ,..,.1. ..., Proc. EPE 93, Brighton, pg. 6-92 [S41] Schumacher , W ., Microprocessor controlled AC servo drive. Micl",II'I ... · I,,'IIII ics in Power Electronics and El ectrical Drives, DaJ'llls l.adl. I !IS:\ 1<'1'1I Fachberichte, pg . 311 [S42J Schum acher, W ., Lconharcl, W., 'T'rilllsistor-fed i\C-!wrvo driv,' wit.h IlIi n o processor CUlltro l. Prol:. lut. Pow,~ r 1·;If'!:L rolli .. ConL, 'I'l) kyo I!m:l, PH. 1·11 IStj:\[ SdJlllllal'h<:r, W., Lc t; ,,~ , 11.- 11 ., L,',"o1l1l)'d, W., Microprtl('(·H:;IIr-cllllt ...,II,·" !II : H""vtl-driv" !4 wit.h H'yn.. hroll" " ~ 11 11" II.II'ylcltr,,"olx IlIOt.ors . I'rol' . 11':1': 1:",, 1'. " " I'"w,')' 1': I,·, ·l.rtllli c:' 11.11.1 Vllr 'n.!" ,· ~;I,, · .. d Ill'ivl'll , 1.,,".1,," 1\IS.'I , p,~ . ·.U:I l~) /I ,ii S\'I IIIII"" 'IIO'I' , W _, M ilII" tI" " 'IIIII" I'.' ·" ·I'.,·II.' ·!' !I ".y", ·lt r"" ~ 1."'IIIIII.I· i "I, . I l !rlll . 'I' \!
11 11" '"1" h W ' ·I~. luSr,
·1,11i
Bibliography
I: ; I ~ , I Schumacher, W ., Rojek, P., Letas, H.H., Hochaufl6sende Lage- und Dreh
)\ahlerfassung optischer Geber für schnelle Stellantriebe. Elektronik, 1985,
S.65 1:;·11'1 Scliumacher, W., Kiel, E., Elektrische Antriebe mit moderner Mikroelek
I.ronik: Der Schlüssel zur Mechatronik. Proc. SPS/IPC/DRIVES 94, Stutt
gilrt, pg . 607-614
I:; 1'(1 Schuster, A., A drive-system with digitally controlled matrix-converter feed
iIlg an AC-induction machine. Proc . Conf. Power electronics and variable
speed drives, IEE 1996, London, pg. 378-382
1:;·1)'1 S<.:efried, E., Stromregelung im Lückbereich von Stromrichter-Gleichstrom
alll.rieben. EJektrie 1976, pg. 308
1:;· I!1j Sceger, G., Selbsteinstellende, modellgestützte Regelung eines Roboters,
I)iss. TU Braunschweig, Vieweg Verlag Braunschweig-Wiesbaden, 1991
I: :!,I) I Serrano-Iribarnegaray, L., The modern space-phasor theory. Part I: Its co IH~ rent formulation and its advantages for transient analysis of converter-fed AC machines. Europ. Trans. on Electr . Power Engng. 1993, pg. 171. Part II: Comparison with the generalised machine theory and the space vector
Lheory. Europ. Trans. on Electr. Power Engng. 1993, pg. 213
I,',!, II Shepherd, W ., On the analysis of three-phase induction motor with voltage
<:o utrol by thyristor switching. IEEE Trans. Ind. Gen. Appl. 1968, pg. 304
I: ;!,:!I Sbepherd, W ., Stanway, l., Slip power reeovery in an induction motor by
I.he use of a thyristor inverter. IEEE Trans. Ind. Appl. 1969, pg. 74
I:·; !,:q Shinohara, K., Mieno, Y., Current control and magnet configurations for higli torque to current ratio in surface mounted permanent magnet syn chrunous motors. Proc. EPE 97 Trondheim, pg. 3.865-3.869 I·';~ "I I Sh l'ivastava, Y., Nguyen, M.T., Sathiakumar, S., Speed estimator for induc I.iull motor. Proc. EPE 99, Lausanne I.' ; ~ ~ I SiIlIOIlS, S., Robuste kartesische Bahnregelung eines sensorgeführten Indus
I.ri(:r<.>I>oters mit digital geregelten Drehstromantrieben. Diss . TU Braun
s(' hw,:ig 1995
I:;!,(\I SillHIJ, D., Hoft, R .G., Microcomputer-controlled single-phase cyclocon
VI" 1., '''' Proc. IECI 1978, pg. 233
I: ;~ ,"I :.; 1
1::""1 ~;"I'" 1-:1<' 1<., Liew, A.C., On-line tuning of rotor flux observers for field
.. ,i'·IlI."d drives usíng improved stator-based flux estimator for low speeds.
1" '0('. I~:I )E 95, Sevilla, pg. 1.437-1.442
1:; loill :·;"1,,,1,1.1\.. , V., Einflufi der Temperatur auf das Betriebsverhalten der dreh
·/.:J dl.:,:regeILen Asynchronmaschine. Siemens Zeitschrift 1969, pg. 760
I: :1iI) I Sorall, r. F., Kisch, D. O., Improved scheme for speed control of an asyn
..h r()Il()U ~ machine by field oriented method. Proc. EPE 91, Firenze, pg. 2-218
1:, 01 1 SI)( OL h, W., Selbstanpassende Regelsysteme in der Antriebstechnik. Diss. TU
I \rarrllschwcig 1971
I:;n:q S"r,:dH:r, .r., David, A., Desdoude, M., High power adjustable speed drive for I" '\\"'" plallL: Current source structure with thyristors for induction machine. I 'ro('.. 1': 1'1': \)5, Scvill;~, pg. 3.056-3.061 1:;I; :1j :-:t., 'ill,nl , 1\ . ,. Co,d,l'ol o [ the illdudioll machinc in Lract.ion . Proc. PEMC 98,
I
I
I: ;1 i'iI
I 'lil/~ , PI\ . l\tl.I -·tl .n ~-;I.""""" , 1',1( ., 1I.,:drrlldll.lll., farrll.- l,ol cntuL COIlLrol ~i'yS I."1Il ror a 'I:~ MW high )'1)(,'1'.1 d,·iv" . l'r,H' . II';I':-.I'ow,'r 1':l<:cl.wlliGll alld Yarinhl<:Sp,'(:d l>riv,::,) l,olldoll IIl HH, "1\ .
:1,1:1
Bibliography
447
[S65] Steiner, M.) Deplazes, R ., Stemmler, H., A new transformerless topology for AC-fed traction vehicles using multi-star induetion motors. Proc. EPE 99) Lausanne [S66] Steinke, l.K .) Switehing frequency optimal PWM control of a three-Ievel
inverter. Proc. EPE 89, Aachen, pg. 1267
[S67] Stemmler) H.) Speisung einer langsam laufenden Synchronmaschine mit
einem direkten Urnrichter. VDE-Fachtagung Elektronik 1969, pg. 177
[S68] Stemmler, H., Ein- und mehrpulsige Unterschwingungswechselrichter, Steu
erverfahren, Strom- und Spannungsverhãltnisse. 1,t IFAC Symp., Control in
Power Electronics and Electrical Drives) Düsseldorf 1974, Vol. 1, pg. 457
[S69] Stemmler, H., Wirk- und Blindleistungsregelung von Netzkupplungsum
formem 50 Hz-16 2/3 Hz mit Umrichterkaskade. BBC-Mitteilungen 1978,
pg. 614
[S70] Stemmler, H., Guggenbach) P.) Configurations of high power voltage source
inverter drives. Proc. EPE 93 , Brighton, pg. 5-7
[S71] Stemmler, H. ) Power electronics in electric traction applieations. Proc.
IECON 93) Maui, pg. 707
[S72] Stephan, R .M.) Americo, M.) Lima, A.G.G., Rodrigues, A., Electric energy conservation in pumping processes at a refinery. Proc. EPE 95, Sevilla) pg. 3.633-3.637 [S73] Stepina) l., Fundamental equations of the space vector analysis of electrical
machines. Acta Technica CSAV 13, 1968, pg. 184
[S74] Stephan, R.M. Field oriented and field acceleration control for induction
motors: is there a difference? Proc. IECON 91, Kobe, pg. 567
[S75] Stoschek, l ., 5 MVA-Umrichter für einen Linearmotorantrieb. BBC Nach
richten 1975, pg. 192
[S76] Strole, D., Vogel) H., TRANSIDYN-Regelungen für Walzwerkantriebe.
Regelungsteehnik 1960, pg , 194
[S77] Strole, D., Adaptivsysteme der elektrischen Antriebstechnik. ETZ-A, 1967,
pg. 182
[S78] SÜti:i, Z., Nagy, 1., lákó, Z., Periodie responses of a nonlinear current con
trolled 1M drive. Proc. EPE 97 Trondheim, pg. 3.847-3.852
[S79] Sugi, K., Naito, Y., Kurosawa, R., Kano, Y., Katayama) S., Yoshida, T.,
A mierocomputer based high eapacity cycloconverter drive for main rolling
mill . Proc. lnt. Power Electr. Conf.) Tokyo 1983, pg. 744
[S80] Sugiura) M., Yamamoto, S., Sawaki, l., Matsuse, K., The basic characteris
tics of two-degrees-of-freedom PlD position controller using a simple design
method for linear servo drives. Proe. Workshop on Advanced Motion Control
(AMC) 96, Tsu City, pg. 59-64
[S81] Sukegawa, T., Kamiyama, K., Mizuno, K., Matsui, T., Okuyama) T. , Fully
digital, vector-controlled PWM VSI-fed AC drives with an inverter dead
time compensation strategy. IEEE Trans. Ind. Appl. 1991, pg. 552
[S82] Sukegawa, T., Kamiyama, K.) Takahashi, l., Imiki, T., Matsutake, M ., A
multiple PWM GTO line-side converter for unity power factor and reduced
harmonics. IEEE Trans. Ind. Appl. 1992, pg. 1302
[S83] Sumner , M., Asher, G.M., Pena, R., The experimental investigation of ro
Lor time constant identification for vector controlled induction motor drives
drrrill~ Lransiellt opetaLing conclition. Proe. EPE 93, Brighton, pg. 5-51
[S84] SV':U: ;:;O Il, '1'., A IILI\ii ll~, M., 'l'lu: JlrodulaLiol1 und control of a matrix converter
• :;.I'IIChrollorrH 1I l/l(' hill" driv, ~ , I')'()(', I':I)E fll, Jo'iwll'l:e, pg. 169-47(. ISH!.I S,i "'h n, M " Y""" IIUI<' Irr., ) 11"1\..111 1" ';/111 , "ill
Bibliography
'I/IX
Tadakuma, S., Ehara, Mo, Ristorical and predicted trends of industrial AC urives. Proc. IECON 93, Maui, pg. 655
I'I'~ I Tadros, Y., Junge, G., Salama, S., Design aspects of high power PWM in verter with IGBT. Proc. EPE 91, Firenze, pg. 2-83
1'I':q Taguchi, T ., Aida, K., Mukai, K., Yanagisawa, T ., Kanai, T., Variable speed
pllmped storage system fed by large-scale cycloconverter. Proc. EPE 91,
Firenze, pg. 2-237
l'I'oq Tajima, R., Rori, Y., Speed sensor-Iess field-orientation control ofthe induc
I.ion machine. IEEE Trans. Ind. Appl. 1993, pg. 175
I'I ' ~ ) I Takahashi, 1., Noguchi, T ., A new quick-response and high efficiency control
~Lrategy of an induction motor. IEEE-Trans. Ind. Appl. 1986, pg.820
I'1'1; I Takahashi, L, Kanmachi, Vltra-wide speed control with a quick torque re
sponse AC servo by a DSP. Proc. EPE 91. Firenze, pg. 3-572
1'1'71 Takeshita, T., Matsui, N., Sensorless control and initia! position estimation
of salient-pole brushless DC motors. Proc. Workshop on Advanced Motion
Control (AMC) 96, Tsu City, pg. 18-23
l'I'xl Tamura, Y., Tanaka, S., Tadakuma, S., Control method and upper limit
of output frequency in circulating current type cycloconverter. Proc. Int.
Semiconductor Power Converter Conf. 1982, pg. 313
1'1'!)1 Taniguchi, K., Tomiyama, Y., Iwatani, T., Irie, R., Acoustic noise reduced
PWM converter and inverter using BJT's and IGBT's. Proc. EPE 91.
l"irenze, pg. 3-252
1'1 '111] 'J'aufiq, J.A., Evaluating the step response of a traction VSI drive. IEE-Power
Electronics and Variable Speed Drives, London 1988, pg. 340
1'1' I I I Tez, E.S., MOTONIC-Chip set: Righ performance intelligent controller for
illdustrial variable speed AC drives. Proc. IEE-Power Electronics and Vari
ahlc Speed Drives, London 1988, pg. 382
1'1' 121 'I'heocharis, J., Petridis, V., Rarmonic insertion in PWM inverter drive
sclternes. Europ. Trans. on Electr. Power Engng. 1992, pg 143
1'1'1:11 'l'hl'llerkauf, R., Digitale Nachbildung drehzahlgeregelter Drehstrom-An
I.ri,'he mit Stromrichterspeisung. IFAC Symp., Control in Power Electronics
illI" I':lcctrical Drives, Düsseldorf 1974, Vol. 1, pg. 709
1'1" ,q ° 1'h'~llerlmllf, H., Zur digitalen Nachbildung von Antriebsscha!tungen mit um
ri ..hL('rgespeisten Asynchronmaschinen. Diss. TU Braunschweig 1975
I'\' I !, I 'l'IIOI"pson, K.R., Acarnley, P.P., Parameter estimation from step excitation 1., '~ l.s 011 a stationary induetion motor. Proc. EPE 97, Trondheim, pg. 4.591
1'1'11
~
':'!JG
I' I ' II; I TiiLillcn, P., Pohjalainen, P., Lalu, J., The next generation motor control
IIldhod: direet torque control (DTC). EPE Journal 1995, no. I, pg. 14-18
1'1' 17 1 'I 'illll"ner, R.A., PWM frequency inverters in the metal industry. Proc. EPE
Firenze, pg. 1-48
W., Cycloconverter drives for rolling mills. IEEE Trans. Ind. Appl.
'l'iIlIJH\ 1'1' I
I !)R2, pg. 400
1'1' ,I) I 'I'hoq~ersen, P., Toennes. M., Jaeger, U., Nielsen, S.E., New high perfor
IIlallCl' vector eontrolled AC drive with automatic energy optimizer. Proc.
1':1'1': !)f" Scvilla, pg. 3.:)81-3.386
I'I' :~()I 'l'IoOlllas, ..1.1,., Boidill, M., An internal model control structure in field ori
'·III."d c.olll.rollcd VSI indllct.ion molors. Proe. F,P]>; 9] . Firenze, pg. 2-202
I' I':! , I 'I'i~ ri-,k, V., Vali:-l, J., II i,~h aCCll r::l(~y and rast respor,,'w digital sp,,('d (f\('<\.Sllrt: 1111'111. 1, ,1' ('",d,!'ol (Ir illdllslrial IIlnl.or driv,:~ . 2"" IFi\ C Sylllp., (:olll.rnl iII l'.,w,' r 1':I,',·l.l'tIllic" fl lld 1'; I"d ricllll)riv"H, i'Jii'lHl'idnrl' 1!J7'7, pg. 7'61 q I.
'x
Bibliography
ti'!!!
[T22] Torok, V., Near optimum on-line modulation of PWM inverters. 3 rd IFAC
Symp. Control in Power Electronics and Electrical Drives, Lausanne 1983,
pg. 247
[T23] Torok, V., Loreth, K., The world's simplest motor for variable speed control?
The Cyrano motor, a PM-biased SR-motor ofhigh torque density. Proc. EPE
93, Brighton, pg. 6-44
[T24] Tomita, M., Sato, M., Yamaguchi, S., Doki, S., Okuma, S., Sensorless esti
mation of rotor position of cylindrical brushless DC motors using eddy cur
rents. Proc. Workshop on Advanced Motion Control (AMC) 96, Tsu City,
pg. 24-28
[T25] Tomizuka, M., Model based prediction, preview and robust controls in mo
tion control systems. Proc. Workshop on Advanced Motion Control (AMC)
96, Tsu City, pg. 1-6
[T26] Tondos, M.S., Minimizing electromechanical oscillations in the drives with
resilient couplings by means of state and disturbance observers. Proc. EPE
93, Brighton, pg. 5-360
[T27] Tondos, M., Mysinski, W., A microprocessor-based speed regulator using state and disturbance observer. Proc. EPE 97, Trondheim, pg. 4.626-4.631 [T28] Toennes, M., Rasmussen, H., Robust self tuning of AC servo drive. Proc.
EPE 91. Firenze, pg. 3-49
[T29] Tripathi, A., Sen, P.C., Comparative Analysis of fixed and sinusoidal band
hysteresis current controllers for voltage source inverters. IEEE Trans, Ind.
EJectr. 1992, pg. 63
[T30] Trondle, H.-P., Regelung elastisch gekoppelter Vielmassensysteme. Siemens
Forschungs- und Entwicklungs-Berichte 1975, pg. 45
[T31] Trondle, R.-P., Regelung einer Spiegelantenne. Siemens Forschungs- und
Entwicklungs- Berichte 1975, pg, 75
[T32] Trzynadlowski, A.M., Kirlin, R.L., Legowski, S., Space vector PWM tech
nique with minimum switching losses and a variable pulse rate. Proc. IECON
93, Maui, pg. 689
[T33] Tso, S.K., Ho, P.T., Dedicated microprocessor scheme for thyristor phase
control of multiphase converters. Proc. IEE 1981, pg. 128
[T34] Tsuji, M., Yamada, E., Parasiliti, F., Tursini, M., A digital parameter iden tification for a vector controlled induction motor. Proc. EPE 97, Trondheim, pg. 4.603-4 .608 [T35] Tsuji, T., Sakakibara, J., Naka, S., CSI drive induction motor by vector
approximation. IEEE Trans. Ind. Appl. 1991, pg.715
[T36] Turnbull, F .G., Selected harmonic reduction in static DC-AC inverters.
IEEE Trans. Comm. Electr. 1964, pg. 374
[T37] Tuttas, C., Pseudo torque control (PTC) - a new control concept for induc
tion motors. Proc. EPE 97 Trondheim, pg. 3.788-3.793
[VI] Vchijima, T., Takigawa, M., Saijo, T., Speed control over a wide operating range of brushless DC motor applying optimal feedback control. Proc. EPE 91. Firenze, pg. 3-54
[V2] Vrbanke, C., Microcomputer control systems for rapid transit vehicles with
AC-drives. Microelectronics in Power Electronics and Electrica! Drives,
Darmstadt 1982, ETG-Fachberichte 11, pg. 401
[ln] ULkill, V.J., Variable structure systems with sliding mode. IEEE Trans. Au 1.0111. enlllr. 1977, pg. 212
IV II van dt:r Hrucck, lL, Álll\l'ysi:-c or !w.rIllollics in volta(,c recl inverter drives
. 1';o.lI s l' d I.,Y I'WM ~ chclJlI'~' witll dj,wolli'.illll(II(:1 swit.chilll\ O] H'rnt.iol I. Pme. EPE
! II , I,' j I' ;1 '1,", I' f!" :1. 'l O I
,I ~',()
IV:2J
Bibliography
van der Broeck, H., Analysis of the voltage harmonics of PWM vol tage fed in
verters using high switching frequencies and different modulation functions.
Europ . Trans. on Electr. Power Engng. 1992, pg. 341
IV :II van der Broeck, H., Gerling, D., Bolte, E., Switched reluctance drive and
PWM induction motor drive compared for low cost applications. Proc. EPE
93, Brighton, pg. 6-71
I V 'II van Haute, S., Henneberger, S., Hameyer, K., Belmans, R., de Temmerman,
.r., de Clerq, J., Design and control of a permanent magnet synchronous motor drive for a hybrid vehicle. Proc. EPE 97 Trondheim, pg. 1.470-1.575 Iv ~ ',I van Wyk, J .D., Holtz, H.R., A simple and reliable four quadrant variable
fr equency AC drive for industrial application up to 50 kW. 2nd IEE Int.
Conf. Electr. Variable Speed Drives, London 1979, pg. 34
IV ('I van Wyck , J .D ., Ferreira, J.A., Evaluation and future prospects of an in tegration technology for hybrid multi-kilowatt power electronic converters. Proc. EPE 97 Trondheim, pg. 1.194-1.199 IV 7] Vas, P., Alaküla, M., Hallenius , K.E., Brown, J.E., Field-oriented control of
saturated AC machines. Proc. IEE-Power Electronics and Variable Speed
Drives, London 1988, pg. 283
IViI] Vas , P., Stronach, A.F., Rashed, M., Zordan, M., Chew, B.C., DSP imple mentation of sensorless DTC induction motor and PM synchronous motor drives with minimized torque ripples. Proc. EPE 99, Lausanne Iv~ll Võlker, C .F., German Patent 906 951, 1944
IV 10] Vollstedt, W., Magyar, P., Adaptive control of DC drives by microcom
puter . Microelectronics in Power Electronics and Electrical Drives, Darm
stadt 1982, ETG-Fachberichte 11, pg. 289
IVIII von Musil, R., Schmatloch, W., Bemessung und Ausführung groBer drehzahl
vcründerbarer Synchronmotoren. Siemens Energie und Automation 1987,
Special: Drehzahlveranderbare elektrische Gro13antriebe, S. 32-41
IVI:l1 Voss, U ., Gleichstromstellertechnik in Nahverkehrsfahrzeugen. Elektrische
Ilithnen 1977, pg. 162
I\lUI Vritnka, P., Griva, G., Profumo, F., Practical improvement of a simple V
I Ilu x estimator for sensorless F.O. controllers operating in the low speed
I,·g im!. Proc. IECON 98, Aachen, pg. 1615-1620
IV I 'ii Vul((lsavic, S.L ., Stojic, M.R., On-line tuning of the rotor time constant for
v<:(' (.o r- controlled induction motor in position control applications. IEEE
' l'r;tIls. llld . Electr . 1993, pg. 130
IWII Wade, S ., Dunnigan, M.W., Williams, B .W ., Improvements for induction IIlitchille vector control. Proc. EPE 95, Sevilla, pg. 1.542-1.546 W;q~ncr , R., Elektronischer Gleichstromsteller rur die Geschwindigkeitss
IW··~ I (.1· IH ,run g elektrischer Triebfahrzeuge. Siemens Zeitschrift 1964, pg. 14
IW :II Wal.;llcr, R, Beitrag zur Theorie des direkten Gleichstrom-Gleichstrom 1IIIII'idlt.ers. Diss. TU Braunschweig 1968
I\
1 ~ )71\, pg. 780
IW !',I WlI.idlllann, M., Drehstromantrieb für Gleichstrombahnen. Siemens Zeit
schrifl l97G, p~. 493
IW(jl WalczYIl<1, A.M ., Koczara, W., Simulation study of current-controlled
dlll lbl y- f(,d illduction machine. Proc. EPE 89, Aachen, pg. 876
IW" I WakY II;l., i\ .M., COlllparison of dyuamics of doubly-fed induction machine , '1III.roll <:d iII lidd- allI' W, ·(·II!;o ·1 ","1 I )1'(·h nl.rlll\l:ly" I.'·II)(·II . I li,,:; . '1'11 II r;tIIlIHChw()il', 1~) 77
Bibliography
451
Walker, L.H., Espelage, P.M., A high performance controlled-current in
verter drive . IEEE Trans. Ind . Appl. 1980, pg. 193
[W10] Wang, Y., Grotstollen, H., Control strategies for the discontinuous current
mode of AC drives with PWM inverters. Proc. EPE 91. Firenze, pg. 3-217
[WllJ Warnecke, O., Einsatz einer doppeltgespeisten Asynchronmaschine in der
groBen Windenergieanlage Growian. Siemens Energietechnik 1983, pg. 364
[W12] Waschatz, U., Adaptive control of electrical drives. Microelectronics in
Power Electronics and Electrical Drives, Darmstadt 1982, ETG-Fachberichte
11, pg. 135
[W13] Waschatz, D., Adaptive Regelung elektrischer Antriebe mit Hilfe von
Mikrorechnern. Diss. TU Braunschweig 1984
[W14] Weh, H., Schrõder, D., Static inverter concepts for multiphase machines with square-wave current-field distributions. Proc. EPE 1985, Brussels, pg. 1.147-1.152 [W15] Weibull , H., Magnusson, T., Valis, J., Standstill testing of properties of
induction motors for inverter control. Proc . EPE 91. Firenze, pg. 2-363
[W16J Weidauer, J., Flexible machines using distributed intelligent drives. Proc.
Workshop on Advanced Motion Control (AMC) 98, Coimbra, pg. 481-486
[W17J Weihrich, G., Drehzahlregelung von Gleichstromantrieben unter Verwen
dung eines Zustands- und StõrgrõBenbeobachters. Regelungstechnik 1978,
pg. 349, 392
[W18] Weihrich, G. , Wohld, D., Adaptive speed control of DC drives using adaptive
observers. Siemens Forschungs- und Entwicklungs-Berichte 1980, pg. 283
[W19J Weiss, H.W., Adjustable speed AC drive systems for pump and compressor
applications. IEEE Trans. Ind. Appl. 1975, pg. 162
[W20] Weiss, H ., Rotor circuit GTO converter for slip-ring induction machine.
Proc. EPE 97 Trondheim, pg. 2.707-2.710
[W21J Weiss, H ., Lampersberger, M ., Control system for the volt age source DC
link converter in the rotor circuit of a slip-ring induction machine. Proc.
EPE 97 Trondheim, pg. 3.853-3.858
[W22] Weiss, H., Ebner, A., Schmidhofer, A., Rotating system tie-frequency con
verter 60 Hz - 25 Hz with rotor converter at a doubly-fed induction machine.
Proc. EPE 99, Lausanne
[W23] Weninger, R, Einflu13 der Maschinenparameter auf Zusatzverluste, Mo
mentoberschwingungen und Kommutierung bei der Umrichterspeisung von
Asynchronmaschinen. AfE 1981, pg. 19
[W24J Wheatley, C.T., Drives. Proc. EPE 93, Brighton, pg. 1-33
[W25] Wheeler, P .W ., Grant, D.A., A low loss matrix converter for AC variable
speed drives. Proc. EPE 93, Brighton, pg. pg. 5-27
[W26] Wieser, R., High dynamic torque calculator for inverter-fed induction ma
chines. Proc. EPE 95, Sevilla, pg. 3.771-3.776
[W27J Wieser, R.S., Kral, Ch., Pirker, F ., Schagginger, M., The Vienna induction
machine monitoring method; On the impact of the field oriented control
structure on real operating behaviour of a faulty machine. Proc. IECON 98,
Aachen, pg. 1544- 1549
[W28] Wilharm, H ., Fick, H ., Einsatz eines Beobachters zur genauen Bestimmung
von Torsionsschwingungen. VDI-Berichte 1978, pg. 69
[W29J Willis, G .H. , A study of the thyratron commutator motor . Gen. Electric
Review 1933
[W:IOJ Wi llller, L., Falk , M., 13cilrag <1m R egelung mechanischer Schwingungen
. Iw illl 11 :~~ p,,J1l VOII llalld. H.<'I',I·III II gs I."ch II iI< 1977, p g. 101
IW:\'I Wd, :"I., M '1'. , Il:irl,',Y, 11.<: ., I lialtll, C ., TIl<' i!.pp li .. al.ioll (Ir IIdd ( ,ri' ~ III., · d
">I .t.I"II." 1.1 ". l'I'I< '. ld, ·tltl 1)( : 11I.,, ·I,ill ". 1" ", . 1': 1' 1'; UI . Fir' ·II ·/.,·, PI',. :\ (i :!' )
[W9]
'1;,2
Bibliography
Bibliography
IW:\2] Wollenberg, G., Schwarz, W., Anwendungen von Mikroprozessoren in nu
merischen Steuerungen für Werkzeugmaschinen. Elektrie 1980, pg. 452
IW:U] Wolters, P., Auslegung stetiger Vorschubantriebe rur Werkzeugmaschinen.
Regelungstechnik 1978, pg. 363
IW :\1 1 Wu, B., Slemon, G.S., Dewan, B.D., PWM-CSI induction motor drive with
phase angle control. IEEE Trans. Ind. Appl. 1991, pg. 970
IW:\;,] Wu, R., Slemon, G.S., A permanent magnet motor drive without a shaft
sensor. IEEE Trans. Ind. Appl. 1991, pg. 1005
IW:\(>! Wu, B., Dewan, S.B., Slemon, G.R., PWM-CSI inverter for induction motor
drives. IEEE Trans. Ind. Appl. 1992, pg. 64
IW:\71 Wu, Y., Fujikawa, K., Kobayashi, H., A torque control method of a two
Jl\ass resonant system with PID-P controller. Proc. Workshop on Advanced
Motion Control (AMC) 98, Coimbra, pg. 240-245
IW:\K] Würslin, R., Pulsumrichtergespeister Asynchronmaschinenantrieb mit hoher Taktfrequenz und sehr groflem Feldschwachbereich. Diss. Univ. Stuttgart, ]984
IW:!!Jl Wyss, W., Gierse, H., Dupré- Latour, B., AC servo drives. Proc. EPE 89,
/\achen, pg. 1061
Xue, Y ., Xu, X., Habetler, T.G ., Divan, D.M., A stator flux-oriented voltage
~o urce variable-speed drive based on DC link measurement. IEEE Trans. Ind.
/\ppl. 1991, pg. 962
I X '2 1 X u, X., Novotny, D.W ., Implementation of direct stator flux orientation
control on a versatile DSP system . IEEE Trans. Ind. Appl. 1991, pg. 694
I x:q Xu, X., Novotny, D.W., Bus utilization of discrete CRPWM inverters for
licld-oriented drives. IEEE Trans. Ind. Appl. 1991, pg. 1128
IX ·II XII , X., Novotny, D.W., Selection ofthe flux reference for induction machine
d"ives in the field weakening region. IEEE Trans. Ind. Appl. 1992, pg. 1353
I Y II YaUIallloto, H., Tomizuka, M., Torsional vibration control for steel mil! drives
via. Itllli-vibration command generation. Proc. IPEC 95, Yokohama, pg. 447
I X II
1 S:!
I\' '.~ I I\': i I
I1''1 1'1' :,1
II, II
Y;Llllada , K., Komada, S., Ishida, M., Hori, T., Analysis ofservo system real
;z,'d hy disturbance observer. Proc. Workshop on Advanced Motion Control
( /\ M(:) %, Tsu City, pg. 338-343
V;lIl(~ , (:., Chin, T.-H., Adaptive speed identification scheme for a vector
"olll.r"IJcd speed sensor less inverter induction motor drive. IEEE Trans.
IlId. /\ppl. 1993, pg. 820
Y" ' "lJ;LIIS, K .A. , The controlled retardation of Ward Leonard drives. Proc.
11':1':, l!J62, pg. 559
Y"o'JUUlS, TCA., Ward Leonard drives, 75 years of development. Electronics
8:. ""wcr, 1968 pg. 144
I.;tch , F .C., Thiel, F.A ., Pulse width modulated inverters for efficiency op
rd
I.illlal control of AC-drives-Switching angles and efficiency jloss profiles. 3 1\0'/\(: Symp. Control in Power Electronics and Electrical Drives, Lausanne
I!JH :\, [lg. 231
I'/: q
1/,:11 1:1.·'1
',I,iil',,'I";II , W. , Drch7.ahlregelung des Asynchronmotors unter Verwendung
,' ; 11<::; B, 'oktchlcrs miL geringer Paramet erempfindlichkeit. Diss. Univ . Er
\;I.III'."II - N ii rllhcr!'" 1984
'/.ai, 1,.< :., I k Marco , ChI' .L., Lipo, 'LA., Ali cx tcnded Kalman filtel' approaeh
to rotor 1.;1111 : <:Il1I,,!.n.llt JtlC1.t6Ur C ln ClIt iJl P\NM illdllc:t.ioll motor drives. IEEE
'I" 'i ll':: . IlId . /\ppl. I\lH'2, pil, . Dr;
'/, /1"""'11 , ,1.1. ., <': l"'(' iIL' <:"rra cllt, /\ ., 1.1"'1 i II 1\ 11
i"cllI (·I. I,," '" 0 !.0/'.
1'1" )('.
lI"il;, :d ('01'111,,1 11.1,;011 li,,· l'O!.or-llllx ('H !.illlll." II':CON il H , /\ 11.<'11<'11,1'1;. :\Il!I. :\1·1
[Z5] [Z6] [Z7] [Z8]
[Z9] [ZI0]
[Z11]
453
Zdenkovic, J., Kuljic, Z., Pasalic, N., Speed sensorless drive with induction machine based on natural field orientation. Proc. EPE 95, Sevilla, pg. 1.752 1.757
Zeitz, M., Nichtlineare Beobachter. Regelungstechnik 1979, pg. 241
Ziegler, M., Hofmann, W., Der Matrix-Umrichter - Kommutierung in nur
zwei Schritten. Proc. SPSjIPCjDRIVES 99, Nürnberg, pg. 521-530
Zimmermann , P., Super-synchronous static converter cascade. 2 nd IFAC
Symp., Control in Power Electronics and Electrical Drives, Düsseldorf 1977,
pg. 559
Zolghadri, M.R., Diallo, D ., Roye, D., Direct torque control system for syn chronous machine. Proc. EPE 97 Trondheim, pg . 3.694-3.699 Zubek, J., Abbondanti, A., Nordby, C., Pulsewidth modulated inverter mo
tor drives with improved modulation. IEEE Trans. Ind. Appl. 1975, pg. 695
Zürneck, H., Ein drehzahlgeregelter spannungsgesteuerter Stromrichter
Asynchronmotor. Diss . TH Darmstadt 1965
Index
Acceleration, 8
- signal, 379
acquisition of signals, 259
active power, 105
actuating variable, 65, 79
actuator, 133, 139
adaptive current control, 137, 144
airgap, 56, 72, 164
alternating current, 103
ampereturns, 51, 166
- -wave, 282
- source, 332
analogue interpolation, 379
angular
- acceleration, 8
frequency, 176
- momentum, 8
- position, 78
velocity, 8, 255
antiparallel thyristors, 143
arc-back, 99
armature, 51
and field control, 56, 61
- circuit, 81
control, 57, 63, 79
- current, 51, 63, 66, 77
- flux, 51
- power supply, 59
- reaction, 52, 60
- reversal, 140
- time constant, 81
- voltage, 58, 66, 79, 90
- winding, 51
asynchronous motor, 163
auxiliary thyristor, 150, 227
Ilack voltal·.p , I ~~ h;~ddll n h , :\ K(J
Iml ll llc 'c'" 1.1". ·.. 1'1 11' ;111
block diagram, 17, 64, 80, 83, 84, 138,
153, 195, 201
blocking voltage, 100
Bode-diagram, 86, 369
brushless DC motor, 341
Cartesian coordinates, 383
cascade control, 81, 83, 89, 90
catenary, 139
ceiling voltage, 91, 96, 262, 265
characteristic equation, 198
circle diagram, 180
circulating current, 144, 238
clock frequency, 223, 276
clutch, 186
coiler, 61
combined fiux model, 300
commutating
capacitor, 226
- coil, 53
- pole, 52
commutation, 52, 103, 123
- failure, 58, 118, 151, 357
- interval, 114, 117
commutator, 51, 163
compensating winding, 52
complex roots, 248
conjugate complex, 87
continuous
- current, 103, 112, 120, 144
- web, 386
control
- amplifier, 83
- band width, 364
- of active/reactive power, 313
of DC Drives, 139
- of synchronous motor, 345
COIILroll('d ()~(~ illator, 248
('or,I.l'ollcl', 7!)
vC)II '(\~" 'Il , '~ II'.~
hll rlc ' 111" ..., I. '/\1 111",,1111 "" 11 11111.1." • . I -IH. ~~IK
willcl
'1( '.
I:\!I
c·"OIvc ·,' I.•". II)H P IIII ' l l\ ll d ('j 1Ild,íIlHt.iIIIl H, ~),I~~
;1;,(i
Index
Index
('onrdinate
sys l.cm, 191, 197
tr:ul sformation, 309
(''' I'rd;tLion, 274
(:""Iolllh friction, 25
I'" "plillt; torque, 369
1'1n::s-over frequency, 85, 135
\ ' III'f<
'nl
I·olll.rol, 81, 85, 135, 136, 275, 304
dlsl.riblltion, 173,174
I"('I IS, 186
1I (1)( ld ,299
plo asor, 177
rl,r( ~ l' enCe , 88
ris(' , 117
snll sor, 143
SO Il rce, 262
converter, 281, 285
v<:cto r, 167, 168, 253
('y<:l"c()lIverter, 304, 343
cy lílltlrical coordinates, 382
1);1.1111'<:1' winding, 331, 342
"" "lpíll t;,86
!';d.í.. , G7, 87, 135
I II:
liJlk , 217, 232
/ 1)( : (,ollverter, 153
d,.J a,,Y, ~7(;
1J,.\ :I,Y('d liríllg, IOG
,J,. II.:< (!) I"",(·l.ioll, 172
dl ·""'I ', lIdísal. ioll , 331
,"'I.III1I 'c1 1I'l x Hlodel, 271 , 274
I I;I ·' ...! ('II/:illl' , 91
tI"I"'''''IU '(' ",!uaL ion, 133
" ilk ... ·II!.i:d cq uaLion, 29,54,195
di ",il.:c.1
",,"Uol, :W7, :$77, 384
::11',lI al pl'l) ce~~ i!lg , 380
"'I',i1.:tI i'ia Líllll , 2Gfí
di",,," , !)7, 100
di r(·(· 1. "OIl'l) () II( 'lIt, :107, 337
ti i:,,·,,"l.iIII10U.S curn'llt, J 1 J , 113 , 144
"i::I. "I'I>:o.II ('(', (;'1, 92, 1:\1
cI,,,",,,\, "'
tllI ''! ,·IIIIVCI'L,·t, 111
cllIl.,v (·v, 'I,· , I 'II)
Il,lI l lI lllI f' I i
1111. 1 \ II (',I 1 ( I I , H~I
"" " ,h"I'., 'nu f ll l '
11 11111 11
I II
II, ~J ( t H
- reference model, 392
Eddy current rotor, 190
efficiency, 185
eigenvalues, 66, 197
electric
- car, 63
- torque, 53
electron tube, 98
electronic switch, 97, 98
encoder, 374, 377
encoderless control, 291
end-effect, 166
energy
conversion, 77
- kinetic, 13, 94
- potential, 13
- storage, 179
equivalent
- circuit, 54, 178, 185, 244
- lag, 85, 381
excitation, 54
extinction margin control, 353
Feed-forward control, 375
feedback
- control, 77
- variable, 82
field
control,58
coordinates, 255, 306, 346
effect transistor, 148, 219
orientated control, 276, 284, 294
orientation, 265
- parameter , 64
reversal, 62
voltage ,91
- weakening, 59, 65, 70, 71 , 90, 258,
276, 336
- winding, 51
filt er, 106
firing, 100
- angle, 107, 112, 120, 137
- circllit, 142
- pulse, 104
flu x
- dellsiLy, 170
- liJlkaGe, 170
IllOdd , '2 :,!), 2(;1, 27 1, 271 , ',li,!} l'I'v""';Jl.I, I,II lIi",II"I:I, :~ I;1l " ,', ' 1."1', 1'{II, 10:1 W l1\lI',
II Y"' I\
111 '/ ,
,111<'111 •
~) nO
:11 /1
flywheel, 34, 94, 323
forced
- commutation, 105, 139, 148, 225 , 231
- cooling, 47
forward blocking, 99
four-quadrant
- drive, 62
- operation, 140
frequency divider, 385
friction, 25
function generator, 108, 244, 276
fundamental component, 103, 132, 231
Gain,80
gas-turbine, 349
gate current, 99, 100
Gate-Turn-Off thyristor (GTO), 98,
148, 236
gear, 11
geometrical progression, 50
gravitational force, 25
gravity, 9
GTO current source converter, 235
Half-controlled converter, 132
Hall sensors, 260
harmonics, 106, 129, 139
heat flow , 44
high dynamic performance, 82
household appliance, 70
Identification, 272
IGBT, 148
- -converter, 219
Ilgner-set , 94
impedance, 77 , 81
impressed
- currents, 245, 304
- voltage, 190
incremental sensor, 266, 378
induced
- armature voltage, 77, 82
- voUage, 53, 62, 72, 90, 91
inductances, 169
induction motor, 163
inertia, 9, 81
inrush cllrrclIL , 190 , 201
insl.a]'ilit,y,27
illsllln(,(:d I·',a!.' : bipola r LrallsiHI.OI'
(I( : Il'l' ), It1 H
i 111" '1',"1" ln; 1.1111 ".I.i ,' , ;~IIH
4 )II/ '. ~ ti
- torque, 349
interphase transformer, 129
inverse
- load model, 364, 376
- transformation, 256
inverter, 117,122
iron losses, 51
isolated neutral, 166, 172, 190, 202
Jerk,393
junction t emperature, 101
Kirchhoffs law, 114
Kramer-drive, 324
Lag element, 66
laminated, 51, 91, 165
latching threshold, 99
lead
- -lag t erm, 83
- term, 135
leakage
- current, 97, 100
- factor , 179, 181
- impedance, 106
line
- -commutated converter, 102, 119,
130, 139, 237
- -side converter, 231 , 355
- -to-line voltages, 120, 203
- -to-neutral voltage, 121, 171, 191
- impedance, 115, 124
- voltage, 103
load
- angle,271
- commutation, 330
- disturbance, 77, 87, 136
- feed-forward, 363
- surge, 63, 89
- torque, 56, 75, 82
- transient, 67
locus of current vector, 285
looper roll, 387
lumped inertia, 174, 363
Machine-side converter, 232, 349
llIagnetic
mllpliri(~r, 78
,·irl'lIil.,71
1i1' I
i It t ~ '1', I II I. i
" '10111 11'.", 1(111
1111 1' 1111 1111 ' 111
1111(.111 " (,11111 , 1, 1
f
1I11 1' ld 't
I 111
457
III HI \ U n II I Hd
HIll
1
III V I ~ , [I
ri, 'r"
158
Index
Index
magnetising current, 180, 246, 276, 284, overflow pulses, 385
293,305
overload capacity, 187
magnetomotive force, 166
overshoot, 87
mathematical model, 174, 276
Park-transformation, 334
Mc Murray circuit, 226
peak torque, 183, 194
IIlcan delay, 137
permanent magnet excitation, 54, 331
IIl cchanical
permeability, 165
model,241
phase-Iocked loop, 385
power,97
phasor , 177
t ime constant, 61, 66, 194
- diagram, 104, 202 , 335
Ill crcury-arc
PI-controller, 83, 92, 135, 137
converter, 78
PID-controller , 84, 135, 267
valve,99
polar coordinates, 281, 383
lTIi croprocessor, 144, 265
position control, 82, 332, 373, 383
Illine hoist, 94 , 343
positive sequence, 202
llIudel of
power
AC machine, 304
- actuator, 97
synchronous motor, 334
- density, 331
IllOdulation, 258
- factor, 106, 131, 181, 187, 351
11I UIl1Cnt of inertia, 8
gain, 107
Illomentum, 7
- tool,70
MOSFET,148
predictive control, 144
IllOlor parameter, 272
preloaded gear, 381
11I oving coordinates, 191
progressive draw, 367
IlII t1li protective interval, 225
loop control, 82, 391
pseudo-random binary sequence, 274
pltasc circuit, 107
pull-out slip, 179, 181
polt: machine, 165
pulse-width modulation, 152, 219, 282
1IIItlI.iplc
pumped storage set, 349
,',ollllDutation, 127
11",tO!' drive, 364
Quadrature axis, 51, 198, 256
11I1I1I,ivilJ'iable,65 quantisation, 265
N:\l II"l'late, 193
11 :11,11r.tI cummutation, 148
1II 'I'" LI,iv,' sequence, 202
II,,:;I.<-d cO lltrol loops, 83 , 374
1II 'ItI,ral 'I.0 Ile, 52
N,'wLo llf; law, 7
IIII·load spced, 59
" " 'lIillal operating point, 70, 181
I"JldillCar feedback, 87
111I11,, :rical
illl,I'I',raLion, 40
ovcl'lIow , 268
(l/'r:<'rv,'J', 2fl!J, 299 , :\71 ()" / Ulr-"o\lLroll,,r , 151\ ,220 01"'11 I""p
1111 , "",,1.1',,1, :~ tl!. IlI lIll"1 I:'II\lc ' I,loll, W{
I..
(' 1" " 11111,,, ,, 1 II "I/dil /I'I, H'L, "
"11 1 ~ "1 \"1l 1l1
('011111 " I, :11 ti
Radius of gyration, 10
railway grid, 303
rare earth permanent magnets, 331
rate-of-change limiter, 88, 364
reactive
- current, 104, 139
- power, 115, 130
recovery time, 101
rectangular components, 339
rectifier, 97, 122
reduced reactive power, 130
reference, 78, 136
frame, 192
- generator, 82, 375
- model, 392
refirill (;, I lH
Sampled data control, 107, 267
sampling
- frequency, 266, 277
- time, 380
saturation, 72, 75, 272
- -dependent flux model, 273
saw-tooth function, 127
Scherbius-drive, 324
screw-down drive, 389
sectional drive, 366
self - excitation, 75
- locking gear, 140
- tuning, 275
sClIIiC:Olldll c.l.or switch, 99
S I~ lI s ill g coi l, 2(iO
:;I'j'
fi: I, ~ 1:1
rc'I:;-l!nc'l'ltl.i" Il, 7H
rC \I~\, (\.'rll Li VI ' ~1I·I.d\,',
residual delay, 80
resolution, 267, 378
resonant circuit, 150, 226
response, 107
restricted speed range, 303
reverse
- current, 102
- voltage, 101
reversing transient, 277
rigid coupling, 81
ring counter, 282
ripple current, 113, 220
rise time, 277, 284
rolling mill, 60, 94, 366, 387
root locus, 65, 198
rotating
- converter, 94, 97
exciter, 95
generator, 78
mass, 97
rotor
coordinates, 335
current vector, 306, 344
- flux, 253
-- coordinates, 345
orientation, 334
- position, 254
- resistance, 272
-- adaptation, 274
- resistor, 175
run-out test, 14
70
. ' ,1 11.(.1 V" li I" t)'I"'''II II1 III, :Ul'f 1 j\lI dt '~ Vllt UI ~ )( t'i rd. ' 11I1 :l, IC J
llc· fÍ C·!j \V !IIIIIII
IH :
1111d.11I \
II " I.! wl ll.! "" '" 111 1
'/ ,1
- wound commutator motor, 163
servo drive, 66, 139, 330, 332
shaft generator, 315
shunting diode, 130, 149
signal
- processing, 259, 265
- processor, 277, 287
simulation, 134
single
- -phase electric brake, 209
- phase converter, 105
six-pulse
- bridge-converter, 141
- converter, 89, 119, 127, 134
slip, 38, 181, 205
- frequency, 178
- power recovery, 323
slot harmonics, 260
smoothing reactor, 113, 126, 137
snubber, 100, 227
solid state, 78
spatial
- distribution, 167, 168
- trajectory, 78
speed
control, 364
- controller, 82, 87, 92
error, 92
- sensor, 369
squirrel cage rotor, 175
stability, 28, 85
stalled torque, 61
standstill, 58
star-connection, 164
starting
- of induction motor, 187
- winding, 189
state space, 390
static converter, 96, 97
stator
- ampereturns- wave, 251
- current vector, 281
- flux
- - coordinates, 290
- - vector, 293
- resistance, 291
- voltage equation, 219
steady state , 57, 107, 177, 204
- mlu:hine model, 242
~Lc p rC'SpOIl::le, (i(i , H7
IÜC "'''1\'' I""p , :lWI
r1 wll,c'II4 ',1 c' tIIlVI'rL1'J':: , IW I1.<'1 d " I',
7~(
459
,li i\1
Index
d,'v ic(', 97
I', "'1 'I(' IICY, 275
I ,al.l...rIl, 229
:: I,;tl.(' , 281
" y",",, 'l.rical """'1)(lIlents, 202
( ;'I'() converter, 286
1I"I'I(",l.ion motor, 190, 195, 197
"pl.iIlIl Jln, 85,135,137
:,wil.ching sequence, 224
; ;,1' """ r,,"ous motor drive, 329
triplen harmoni cs, 227
turn-off, 100
twin
- drive, 381
- motor, 94
two-
mass system, 370
phase oscillator, 247
pulse converter, 119, 137
quadrant operation, 106
types of AC machines, 175
'I':II'I ''' "lc ter generator, 78
1."I', "i" a l voltage, 234, 262
I.llf'f ll"d
,;tp,tcil.y, 100
\,,. ;o.lI sicnt, 43, 46
Unidirectional torque, 139
universal motor, 70
unsymmetrical
starting circuit, 211
- supply voltages, 203
i.I"" "- I>i,,~~e
I'""verter, 11 9
::< 'I'l'l y, 175
1.I',\' l'Ill.roll, 99,329
1,lo y ,'isl.l)r, ~7, 100, 103, 107
(,lI lIv( ~ rLe r, 225
", "
1.''''1 \,«' , S, 178,193
,' ,,"Lrol, 81, 154
1'"liL, 77
1., ''''1I1 I'/~ pccd
" III' VI'S, Gil, 243
plnlll " r,!), 78, 147
\.r /Ii 'h, 'd v" hicl cs, 388
I "1<'1,;"" dri ve, 69,130
1.1 11 11 ,,, 1111 ' 1'1',
1' 111,, :1'1'1' :\11 1\
il:\
Vectorial PWM, 221, 277
vehicle drive, 73
volt age
-time-area, 116, 127
controlled oscillator, 385
feed-forward, 312
gain, 137
limiter, 293
model,298
source converter, 218, 275
vector, 177, 193
Waiting interval , 133
Ward Leonard, 78, 85, 93
waveform, 166
rllll cLioIl, 66, 83, 84, 138, 268,
I.IIlII :: l'm,"al.ioll, 195, 254, 339
\ ' " " 'dl '"L , (i", 87, 136, 199
IIIII', '
Zero
current interval, 144
sequence, 202
vector, 223