flux miller 2

8/9/2019 Flux Miller 2

1/10

Staton, D.A. and Deodhar, R.P. and Soong, W.L. and Miller, T.J.E.(1996) Torque prediction using the flux-MMF diagram in AC, DC, and

reluctance motors.IEEE Transactions on Industry Applications

32(1):pp. 180-188.

http://eprints.gla.ac.uk/archive/00002844/


2/10

180 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 32, NO. 1, JANUARYFEBRUARY 1996

ue Prediction Using the Flux-MMFin AC, DC, and Reluctance MotorsD a v i d A. Sta ton , Ra je sh P. Deodhar, Student Member, IEEE, W e n L. S o o n g , Member, IEEE,

and Timothy J. E. Miller, Senior Member, IEEE

Abstract-This paper uses the flux-MMF diagram to compareand contrast the torque production mechanism in seven commontypes of electric motor. The flux-mmf diagram is a generakedversion of the flux-linkage versus current (+i) diagram orswitched-reluctance motors. It is illustrated or switched-reluctance, synchronous-reluctance, induction, brushless ac,brushless dc, interior PM and dc commutator motors. Thecalculated flux-MMF diagrams for motors with the sameelectromagnetic volume, airgap, slotfill, and total copper loss areshown and are used to compare the low-speed torque and torqueripple performance. The motor designs used were reasonablyoptimized using a combination of commercially available motorCA D packages and finite-element analysis.

1. INTRODUCTIONHE FLUX-LINKAGE versus current ($4) iagram hastraditionally been used to analyze the switched-reluctancemotor (SRM) performance characteristics [11-[6]. Recentlythis concept has been extended to synchronous-reluctance(SYNCHREL) m otors [5 ] and doubly salient permanent mag-net (DSPM) motors [6] in order to facilitate a direct compar-ison with the SRM. The concept of electromagnetic torquebeing derived from the ( $ - a ) characteristics is howev er funda-men tal to all electric motors [7]-[9]. This paper shows how theflux-MM F diagram can be derived from the ($4)iagram and

applied to several motor types, providing a unified method ofanalysis and evaluation.Section I1 describes the salient features of the flux-MMFdiagram and gives the relevant definitions of flux and M MF.Section 111 explains the constraints applied in designing theseven motor types used in the comparison. Section IV de -scribes the flux-MM F diagram of each m otor. Finally SectionV compares and contrasts the seven motor types.11. THEFLUX-MMFDIAGRAM

Th e $-i diagram is based on plotting the variation ofinstantaneous phase flux-linkage against the instantaneousphase current. The instantaneous torque per phase is givenPaper IPCSD 95-53, approved by the Electric Machines Committee ofthe IEEE Industry Applications Society for presentation at the 1994 IndustryApplications Society Annual Meeting, Denver, CO, October 2-7. Manuscrip treleased for publication July 11, 1995.D. A. Staton is with Control Techniques plc, Newtown, Wales SY16 3AJ,U.K.R. P. Deodhar and T. J. E. Miller are with the SPEED Laboratory,University of Glasgow, U.K.W . L. Soong is with the GE Corporate Research and Development,Schenectady, NY 12301 USA.Publisher Item Identifier S 0093-9994(96)00322-2.

where W is the eo-energy, a is the instantaneous phasecurrent and B is the rotor position. The partial differentiationis performed by dividing the area enclosed between twosuccessive magnetization curves by the rotor displacement.The average electromagnetic torque per phase is proportionalto the area enclosed by the $-i locus over one electrical cycle.Thus mPTAV =- W2nwhere W is the energy converted as shown in Fig. 1, m isthe number of phases and p is the number of pole-pairs.

To eliminate the arbitrariness of the number of turns fromthe an alysis, the phase flux-linkage and current are respectivelydivided and multiplied by Nph, he series turndphase, to givethe effective phase flux and phase M MF . These are sim ilar toper-unit quantities. The quantity $/Nph is not the actual flux inthe machine but the e f fec t i ve$u which links the winding andthus produces torque. The con verted energy can be calculatedfrom either the flux-MMF or $-z locus, since both enclosethe same area as shown in Fig. 1. The flux-MMF diagramis independent of the particular voltage level for which themotor happens to be wound. The important features of theflux-MMF diagram are the following:

Area enclosed indicates torque capability.* Shape indicates nature of excitation (ideally ellipsoidalfor sinewave and rectangular for squarewave).Deviation from ideal shape and uneven spacing of mag-netization curves indicates torque ripple and saturation.Whereas d- q axis theory can only predict average torque,this method can p redict torque ripple as well.Orientation of the major axis with respect to the MMF-axis indicates the kVA utilization (power factor insinewave motors).Electric and magnetic loadings and their utilization areindicated graphically.The flux-MMF diagram for a particular motor is shownwith magnetization curves superimposed (e.g., Figs. 5an d 12).These are generated using a series of static finite-elementanalyses at incremental rotor positions ( e ) and increasingloads.The method of predicting torque ripple using the $-i dia-gram has been validated for synchronous reluctance motors

0093-9994/96$05.00 0 996 IEEE


3/10

/ , ' ' .:.. '.'

OS-0%irgapSlot fill

......

. . . .. ... .. . . .

I

I 1

copper loss

Fig. 1 . The equivalence between the +-z and flux-MMF diagrams.

634W

TABLE IDESIGNCONSTRAINTS

8 inches (203.2mm)

with measurements as shown in Figs. 5and 6 [ 5 ] . In theinterest of simplicity this paper presents only the electromag-netic torque and neglects cogging, which contributes no usefulaverage torque. How ever, it is possible to calculate the coggingtorque using the flux-MMF diagram technique [101. Briefly,this is done by calculating the variation in magnet workingpoint w ith rotation. The flux-MM F diagram is then constructedusing the demagnetization characteristic.

111. DESIGNCONSTRAINTSA standard D132 frame induction motor (Section IV-A) hasbeen used as the basis for designing all the motors used inthis comparison. Each has the same electromagnetic volume,airgap, slot-fill, and total copper loss (at 115C an d 7.5 kW

rating) as the induction motor (Table I). Friction, windage, andiron losses are not quoted here, so that the conclusions madefrom the com parison are strictly valid at low speed.Iv. MOTORSUSED IN THE COMPARISON

A. Induction Motor (ZM)This is a standard D132 induction motor, capable of 11 kWwhen operated from a 50-Hz sinusoidal supply. It must be

15 .- - - - . SPM-AC I

-15-3 -2 -1MMF [U-t]

Fig. 2. flux-MMF diagram for the IM , SYNCHREL and SPM-AC motors(full-load).

STATON et al.: TORQUE PREDICTION USING THE FLUX-MMF DIAGRAM IN AC, DC, AND RELUCTANCE MOTORS 181

derated to 7.5 kW for operation from a variable speed supply it has been shown that this construction gives the maximum

Fig. 3. SYNCHREL motor design.

(10 : 1 speed ratio) with constant torque (50 Nm). The totalcopper loss at the 7.5 kW rating is equal to 634 W (statorcopper loss = 426 W ; rotor copper loss = 20 8 W); this is themain design constraint used in the com parison.The flux-MMF diagram for the IM is derived from thephasor d iagram. The instantaneous voltage and current is givenby

i = i c o s (u t- 4); 'U = v c o s (ut).The flux-linkage is calculated using

X = J' v d t = V / w in (ut) .The flux-MMF diagram for the IM at full-load is showntogether with that of the SYNCHREL (Section IV-B) andSPM-AC (Section IV-D) motors in Fig. 2.B. Synchronous Reluctance Motor (SYNCHREL)

An axially laminated rotor is used in the comparison since


4/10

182 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS,VOL. 32, NO 1, JANUARYIFEBRUARY 1996

70MEASURED - * FE ANALYSIS

Iph =5A!0 U) 40 60 80 100 120 140 160 180ANGLE [ELECTRICALDEGREES]

Fig. 6 . SYNCHREL measured and calculated (FE analysis) torque-ripple.

MM F [ U- t ] Fig. 7. SRM designFi g 5motor (6 phase current values with the maximum at full-load)Flux-MM F diagram and m agnetization curves for the SYNCHREL as shown in Fig. 8.Fig. 9compares the flux-MMF diagram atfull-load with that of the SYNCHREL motor (Section V-B).saliency ratio [11]-[20]. The stator and winding type are thesame as used in the induction motor. The motor cross sectionis shown in Fig. 3. This is a screen-capture taken from asynchronous reluctance motor design program (PC-SREL) .Anon-load flux plot taken from the finite-element analysis isshown in Fig. 4. The predicted flux-MM F diagram is shown inFig. 5and the torque waveform in Fig. 6.The angle y etweenthe stator MMF axis and the rotos q-axis (high-inductance axis)is set at 60' to give maximum torque/ampere under saturatedconditions [121.C. Switched Reluctance Motor ( S R M )

The calculation of SRM torque from the $- a locus is welldocumented [I]-[3]. The 6 :4 SRM used in this analysis wasdesigned using the PC-SRD program [l], [3], [22] and isshown in Fig. 7.It has the same number of phases and energyconversion strokes per revolution as the other brushless motorsused in the comparison. The 6 :4 motor has high torque ripple

D. Suifiuce Permanent Magn et Brushless AC (SPM-A C)The surface permanent magnet brushless ac (SPM-AC)motor used in the analysis is shown in Fig. 10. It was de-signed using the PC-BDC program [21], [22]. It has rare-earthmagnets with a remanence of 1.1 T to give flux-density levelscomparable to those in the IM. The magnet arc and distributedwinding configuration were selected to give a sinusoidal back-emf.Fig. 12 plots the flux-MMF diagram (with magnetizationcurves) at 6 currents ranging from no-load to full-load. Thedynamic locus of the operating point (whose coordinates areM M F and effective flux) is nearly elliptical. The magnetizationcurves are virtually straight because the flux-linkage is setby the magnets and is virtually independent of the current.Slight irregularities in the otherwise regular spacing of themagnetization curves indicate a small amount of torque ripple


5/10

STATON et al.: TORQUE PREDICTION USING THE FLUX-MMF DIAGRAM IN AC, DC, AND RELUCTANCE MOTORS

10

5 -

-x3y.-5

-10

I4. 0 6.00 8.00 10.00 11.00 14.00 16.00

.2.00Rotor posmon ( deq ) x 1 . 0 ~ 1

-

-

-

183

Fig. 8. SRM phase current and torque waveforms (150 rhin).Fig. 11. SPM-AC flux plot (full-load, 8 = Z O O )

SRMSYNCHREL 6OdegCURRENTANGLE)

- - - - - - -

-100 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5MMF [kA-I]

Fig. 9. Flux-MMF diagram for the SYNCHREL and SRM

Fig. 10. SPM-AC and SPM-DCl motor designs.

due to slotting (Fig. 13). A typical full-load flux plot takenfrom the finite-element analysis is shown in Fig. 11.

15SPM-AC SURFACE PERMANENTMAGNET BRUSHLESS AC MOTOR

-15 1-3 -2 1 0 1 2MMF [M-t]

Fig. 12. flux-MMF diagram and magnetization curves for the SPM-ACmotor (6 phase current values with the maximum at full-load).

- PM-AC SURFACEPM BRUSHLESS A C MOTORSPM-DC1: SURFACEPM BRUSHLESSDC MOTORBOTH MOTORS HAVEA 150 DEGREEMAGNET ARCz mi - - - - -* O I00 20 40 60 80 100 120 140 160 180

ANGLE [ELECTRICALDEGREES]Fig. 13. SPM-AC and SPM-DC1 torque ripple (FE predication at full-load).


6/10

184 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL 32, NO . 1, JANUARYiFFBRUARY 1996

15 SPM-DC1: SURFACE PERMANENTWGNET BRUSHLESS DC MOTOR

-15 1 -2 1 0 1 2 3MMF [kA-t]

Fig. 14.motor (8 phase current values with the maximum at full-load).Flux-MMF diagram and magnetization curves for the SPM-DCI

1

101- PM-DC2: BRUSWLESSDCMOTOR (17Odeg MAGNETARC)I I

'0 20 40 60 80 100 1u) 140 160 I80ANGLE [ELECTRICAL DEGREES]

Fig. 15. SPM-D CI and SPM -DC2 torque ripple (FE prediction at full-load).

Fig. 16. IPM motor design.

E. S uf ac e Permanent Magnet Brushless DC (SP M-DC )Tw o surface permanent magnet brushless dc (SPM-DC)motors were designed to operate with 120" squarewave linecurrents:1) SPM-DC1 with a 150" magnet arc: this uses the samegeometry and magnets as the SPM-AC motor, witha concentrated winding to give a flat-top back-emfwaveform.2) SPM -DC 2 with a 17 0" magnet arc: this motor is exactlythe same as SP M- DC 1 except that it uses a larger magnet

arc to reduce the torque ripple.Fig. 14is the flux-MMF diagram for the S PM-D C1 motor.The diagram has a nearly ideal rectangular shape and regularspacing of the magnetization curves, with a small amount oftorque ripple that is reduced in the SPM -DC 2 design (Fig. 15).F . Interior Permanent Magnet Motor (IPM)

The interior permanent magnet (IPM) motor was also de-signed using PC-BDC,Fig. 16. It was designed to operate fro ma sinusoidal supply and uses the same stator and winding as

Fig. 17. P M lux plot (full-load, 0 = 20').

the SPM-AC motor. Fig. 18 is the flux-MMF diagram (withy = 30" selected for maximum torque/ampere). The contrastbetween flux-MMF diagrams for the SPM -AC and IPM motorsis evident from Figs. 12 and 18, respectively. A lthough bothmotors have the same average torque, the deviation fromthe near ideal elliptical shape and uneven spacing of themagnetization curves indicate larger torque ripple in the IP M(Fig. 19). A typical full-load flux plot taken from the finite-element analysis is shown in Fig. 17.G. Permanent Magnet DC Commutator Motor (PM DC )

The permanent magnet dc commutator motor (PMDC) usedin the analysis is shown in Fig. 20. It was designed using thePC-DCM program [23]. It has the same numbers of slots andpoles as the SPM-DC motor, resulting in a "9 phase" motor.The m agnet has a remanence of 0.75 T. This is less than the1.1T used in the SP M-D C motor but is required to increase thecopper space, noting the flux-concentration in PMDC motors.The flux-MMF diagram is shown in Fig. 22 and the predicted


7/10


4-POL E SPOKE MOTOR WITH A 30 DEGREE CURRENTANGLE-3 -2 1 0 1 2 3-15

185

in

5 -f-X2

5 -

-10

-

-

- PMAC PM BRUSHLESS AC MOTORIPM SPOK E MOTORWITH 0 DEGREE CURRENTANGLE

0 20 40 60 80 100 120 140 160 180ANGLE [ELECTRICAL DEGREES]

Fi g 19 SPM-AC and IPM torque ripple (FE prediction at full-load)

Fig. 20. PMDC motor design.

torque ripple in Fig. 23. A typical full-load flux plot takenfrom the finite-element analysis is shown in Fig. 21.

Fig. 21. PMDC flux plot (full-load, 0 = loo).

15 I

-15 I-0.6 -0.4 -0.2 0 0.2 0. 4 0.6MMF [kA-t]

Fig. 22.(6 values of current with the maximum at full-load).Flux-M MF diagram and magnetization curves for the PMDC motor

I 1

*O t 10 20 40 M) 80 100 1u) 140 160 180

ANGLE [ELEClWCAL DEGREES]Fig. 23. Torque ripple in the PMDC and SPM-DC2 motors (full-load).


8/10

186 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 32, NO. 1, JANUARYEEBRUARY 1996

15 I

-151 I-3 -2 1 0 1 2 3MMF [M-t]

Fig. 24. Flux-MMF diagram for th e SPM-AC and SPM-DCl motors(full-load).

-- I

I5t!

-15-3 -2 -1 0 1 2 3

Il 0 l

5!


9/10


5 -

4 -7=kk52 3 -Y::2-w

1 -

187

0-

TORQUE1001

50waw

2010 IM IRE LIRM

Fig. 26. Average rated torque in the 7 motors used in the comparison.TORQUE RIPPLE801

. . . . . . . . . . .

. . . . . . . . . . . . . .w . . . . . . . . . . . . . .

. . . . . . . . . . . -.......................

. . . . . . . .

.........................

~~ ~IM SREL SRM SPMAC SPMDC IPM PMDCFig. 27. Torque ripple in the 7 motors used in the comparison.

E. SPM-DC an d PMDCThe comparison between the SPM-DC and PMDC motorsis complicated by the fact that they have different number ofphases. This results in a low peak phase MMF in the PMDCmotor. Even if the phase ratio (9/3) is used to scale MMFs,the PMDC motor has a smaller MMF. This is mainly due tothe fact that the SPM-DC motor has its slots on the outsideof the motor and has a greater winding area. Even though alower grade magnet is used in the PM DC motor, the effective

flux is the same as in the SPM -DC motor. This is mainly dueto the larger rotor diameter in the PMDC motor.The PMDC motor may need to be derated with respect tothe other motors because most of the losses are generated inthe armature which is not as easily cooled as when it is onthe outside of the motor.VI. CONCLUSIONS

The flux-MMF diagram is a general theoretical meansfor calculating and comparing the average and instantaneous

FLUX

1RELFig. 28. Effective flux in the 7 motors used in the comparison.

MMF-7-7

IM 1PMAC IPMIMDCFig. 29. MMF in the 7 motors used in the comparison.torque of a range of electric motors of different types. Themethod goes far beyond the capability of unified or gen-eralized machine theory:It can deal with motors that cannot be analyzed usingtraditional methods for example, the doubly-salient SRMand the squarewave brushless dc motor.It can calculate instantaneous as well as average torque.It can calculate cogging and electromagnetic torque [lo].It makes direct use of finite-element calculations in a waythat is not possible with generalized-machine or lumped-parameter theories, and presents many of its results viagraphical images whose features are readily associatedwith particular characteristics such as torque per unitvolume, kVA requirement, torque ripple, etc.Because it is based on co-energy calculations and non-linear magnetization data, the method is not restricted byassumptions of linearity or sinusoidal MMF distributions.

It would be unwise to proclaim the method as a modern-day replacement for generalized machine theory, which will


10/10

188 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 32, NO. 1, JANUARYFEBRUARY 1996

surely always be the foundation theory for ac and dc machinesas well as fo r field-oriented control and some aspects ofpower-systems analysis; but it should help in the analysis andcomparative evaluation of a wide range of electrical machinesincluding new arrivals that fit no classical model.REFERENCES

[ l ] T. J .E. Miller, Switched Reluctance Motors and T heir Control. Oxford:Magna Physics Publishing and Oxford University Press, 1993.[2] __ Brushless Perm anent-M agnet and Reluctance Motor Drive s.Oxford: Oxford University Press, 1989.[3] T. J. E. Miller and M. I. McGilp, Nonlinear theory of the switchedreluctance motor for rapid computer-aided design, in IE E Proc.-E, vol.137, no. 6, Nov. 1990.[4] F. Liang, Y. Liao, and T. A. Lipo, A new variable reluctance motorutilizing an auxiliary commutation w inding, IEEE Trans. Znd. Applicat.,vol. 30, no. 2, March/April 1994.[5] D. A. Staton, W. L. Soong, and T. J. E. Miller, Unified theory of torqueproduction in switched reluctance and synchronous reluctance motors,IEEE Trans. Ind. Applicat., vol. 31, no. 2, MarcMApril 1995.[6] Y. Liao, F. Liang, and T. A. Lipo, A novel permanent magnet motorwith doubly salient structure, in Con$ Rec. IEEE-IAS Annu. Meeting,Houston, Oct. 1992.[7] H. H. Woodson and E. Melchen, Electromechanical Dynamics. Ne wYork: Wiley, 1968.[8] D. C. White and H. H. Woodson, Electromechanical Energy Conversion.New York: Wiley, 1959.[9] A. E. Fitzgerald and J. R. Kingsley, Electrical Machinery, 2nd Ed.New York: McGraw-Hill, 1961.[ l o ] R. P. Deodhar, D. A. Staton, T. M. Jahns, and T. J. E. Miller, Predictionof cogging torque using the flux-MM F diagram technique, in Con5 Rec.IEEE-IAS Annu. Meeting, Orlando, FL, Oct. 1995.

[111 D. A. Staton , T. J. E. Miller, and S. E. Wood, Maximising the saliencyratio of the synchronous reluctance motor, in IEE Proc. B, vol. 140,no.4, July 1993, pp. 249-259.[12] L. Xu , X. Xu , T. A. Lipo, and D. W. Novotny, Control of a synchronousreluctance motor including saturation and iron loss, IEEE Trans. Znd.Applicat. , vol. 27, no. 5 , pp. 977-985, Sept/Oct. 1991.[13] I. Boldea, Z. X. Fu, and S. . Nasar, Performance evaluation of axiallylaminated anisotropic (ALA) rotor reluctance synchronous motors,IEEE Trans. Ind. Applicat., vol. 30, no. 4, July/August 1994.1141 A. J. 0. Cruickshank, A. F. Anderson, and R. W. Menzies, Theory andperformances of reluctance motors with axially laminated anisotropicrotors, in Proc. IEE, vol. 118, no. 7, 1971.D. Platt, Reluctance motor with strong anisotropy, in Cont Rec.IEEE-IAS Annu. Meeting, Seattle, WA, Oct. 1990, pp. 225-229.A. Fratta and A. Vagati, Axially laminated reluctance motor: Ananalytical approach to the magnetic behaviour, in Int. Cont Elec.Mach., Pisa, Italy, vol. 111, Sept. 1988, pp. 1-6.T. A. Lipo, A. Vagati, L. Malesani and T. Fukao, Synchronousreluctance motors and drives-A new alternative, in IEEE-IAS Annu.Meeting, Houston, Oct. 1992.A. Vagati and T. A. Lipo, Synchronous reluctance motors anddrives-A new alternative, in IEEE-IAS A nnu. Meeting, Denver, Oct.1994.1191 R. E. Betz and T. J. E. Miller, Aspects of the control of synchronousreluctance machines, in Euroaean Power Electron. Conf, EPEYI ,Florence, Sept. 1991.W. L. Soong, D. A. Staton, and T. J. E. Miller, Design of a new-axially-laminated interior permanent magnet motor, IEEE Trans. Ind.Applicat. , vol. 31, no. 2, MarcWApril 1995.J. R. Hendershot and T. J. E. Miller, Design of brushless permanent-magnet motors. Oxford: Magna Physics Publishing and Oxford Uni-versity Press, 1994.D. A. Staton, M. I. McGilp, T. J. E. Miller, and G. Gray, High-speedPC-based CAD for motor drives, in 5th EPE CO$, Brighton, Sept.1993.D. A. Staton, T. J. E. Miller, and M. I. McGilp, Interactive computeraided design of permanent magnet D C motors, in Con$ Rec. IEEE-IASAnnu. Meeting, Toronto, Canada, Oct. 1993.R. P. Deodhar, D. A. Staton, and T. J. E. Miller, Modelling of skewusing the flux-MMF diagram, IEEE-PEDES Con$, New Delhi, India,Jan. 1996, in press.

David A. Staton was born in Chesterfield, England,on July 29, 1961 He received the B Sc (Hons )degree in electrical and electronic engineeringfrom Trent Polytechnic, Nottingham, England, in1983 and the Ph D degree from the University ofSheffield, England, in 1988From 1977 to 1984, he was employed byBntish Coal, who sponsored him while he wasundertakmg the B Sc degree While at theUniversity of Sheffield, he developed C AD softwarefor permanent-magnet dc motors in collaborationwith GEC Electromotors Ltd From 1 988 to 1989, he was with Thorn EM1Central Research Lahoratones and was engaged in the design of motorsfor the Kenwood range of food processors From 1989 to 1995, he wasemployed as a research assistant in the SPEED Laboratory at the Universityof Glasgow Presently he is with Control Techniques plc, Newtown, WalesHis research interests are in the computer aided design of permanent-magnetand reluctance motors

Rajesh P. Deodhar (S88-M90-S94) was born inBombay, India, on February 22, 1968. He receivedthe Bachelor of Engineering (B.Eng.) degree in elec-tronics engineering from the University of Bombayin 1989 and the Master of Technology (M.Tech.)degree in electronic product design from the Centerfor Electronic Design and Technology (CEDT) atthe Indian Institute of Science, Bangalore, in 1991.Since the beginning of 1994, he has been workingtowards the Ph.D. degre e in the SPEED L aboratory,University of Glasg ow, Scotland, in the generalizedarea of design and analysis of electric machines and drives.From 1991 to 1993, he worked as a Design Engineer with CromptonGreaves Ltd., Bombay, on the design of induction and brushless PM motorsand drives. In 1993 he spent six months in Japan, working with Hitachi Ltd.,on the applications of electric machines and drives in home appliances.Mr. Deodhar was awarded the CEDT Design Gold Medal by the IndianInstitute of Science in 1991. He also received the Hitachi-AOTS scholarshipin 1993 which enabled him to w ork with Hitachi Ltd. in Japan for six months.H e is an Associate Member of the IEE, U.K.

Wen L. Soong (S89-M90-S90-M93) was bornin Kuala Lampur, Malayasia, and received theB Eng degree from Adelaide University, Australia,In 1989, and the Ph D degree from GlasgowUniversity, Scotland, in 1993He is an electrical engineer in the Power ControlsPrograms at General Electric Corporate Researchand Development, Schenectady, NY His researchinterests include permanent magnet and reluctancemotor design and modeling, field-weakeningcontrol, and sensorless control

Timothy J. E. Miller (M74-SM82) is a native ofWigan, U K He was educated at Atlantic College,the University of Glasgow, Scotland, and the Uni-versity of LeedsFrom 1979 to 1986 he was an electrical engineerand program manager at General Electric CorporateResearch and Development, Schenectady, NY Hisindustnal experience includes periods with GEC(U K ), British Gas, Intemational Research and De-velopment, and a student-apprenticeship with TubeInvestments Ltd He is Lucas Professor in Powe rElectronics, and founder and Director of the SPEED Consortium at theUniversity of Glasgow, Scotland He is the author of 105 publications inthe fields of motors, dnves, power systems, and power electronicsProf Miller is a Fellow of the Royal Society of Edinburgh and a Fellowof the IEE

flux miller 2

Documents