IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO.

3, MAY 2006

711

DTFC-SVM Motion-Sensorless Control of a PM-Assisted Reluctance Synchronous Machine as Starter-Alternator for Hybrid Electric Vehicles
Ion Boldea, Fellow, IEEE, Cristian Ilie Pitic, Cristian Lascu, Gheorghe-Daniel Andreescu, Senior Member, IEEE, Lucian Tutelea, Frede Blaabjerg, Fellow, IEEE, and Per Sandholdt, Member, IEEE

AbstractPermanent magnet-assisted reluctance synchronous machine (PMRSM) starter alternator systems are credited with good performance for wide speed range in hybrid electric vehicles. This paper proposes a motion-sensorless motor/generator control of PMRSM from zero speed up to maximum speed, using direct torque and ux control with space vector modulation. A quasioptimal stator ux reference with a ux versus torque functional is proposed. A stator ux observer in wide speed range uses combined voltage-current models for low speeds, and only the voltage model for medium to high speeds, both in proportional-integral closed loop. A novel rotor speed and position observer with a fusion strategy employs signal injection and only one D-module vector lter in stator referencefor low speed, combined with a speed observer from the stator ux vector estimationfor medium-high speed. The proposed system is introduced piece by piece and then implemented on a dSpace 1103 control board with a 350-A metal-oxide-semiconductor eld-effect transistor-pulse-width modulation converter connected to a 42-Vdc, 55-Ah battery, and a 140-Nm peak torque PMRSM. Extensive experimental results from very low speed to high speed, regarding observers and drive responses, including articial loading (motoring and generating), seem very encouraging for future starter-alternator systems. Index TermsDirect torque and ux control (DTFC), hybrid electric vehicles (HEV), integrated starter-alternators (ISA), permanent magnet-assisted reluctance synchronous motor (PM-RSM), sensorless control.

I. INTRODUCTION HE integrated starter-alternators (ISA) topic represents a hot subject for internal combustion engines (ICE) and hybrid electric vehicles (HEV). Typical specications for ISAs de[1] pend on the degree of vehicle electrication

peak electric power peak electric power peak ICE power

(1)

Manuscript received March 8, 2005; revised October 26, 2005. Recommended by Associate Editor J. Shen. I. Boldea, C. I. Pitic, C. Lascu, and L. Tutelea are with the Department of Electrical Machines and Drives, University Politehnica of Timisoara, Timisoara 300223, Romania (e-mail: boldea@lselinux.upt.ro; pitic.c@lselinux.upt.ro; cristi@et.upt.ro; luci@lselinux.upt.ro). G.-D. Andreescu is with the Department of Automation and Applied Informatics, University Politehnica of Timisoara,Timisoara 300223, Romania (e-mail: daniel.andreescu@aut.upt.ro). F. Blaabjerg is with Institute of Energy Technology, Aalborg University, Aalborg East 9220, Denmark (e-mail: fbl@iet.aau.dk). P. Sandholdt is with Sauer-Danfoss ApS, Nordborg 6430, Denmark (e-mail: psandholdt@sauer-danfoss.dk). Digital Object Identier 10.1109/TPEL.2006.872369

varies from less then 10% in conventional ICE vehicles to 1025% in mild hybrid vehicles, 2550% in full parallel hybrid vehicles, and 100% in series hybrid and fully electric vehicles. In all cases a wide constant power-speed range (from 4:1 to 12:1) is benecial. From this point of view, corroborated with low inverter kVA and low total system losses, the permanent magnet reluctance synchronous motor (PMRSM) seems to be one of the best [2], together with the electrically dc-excited claw-pole [3], or standard rotor synchronous machines [4]. Quite a few research teams have recently investigated in depth the design, performance [5][8] and vector-control [5] of PMRSM. Sensorless vector-control for interior permanent magnet synchronous motors (IPMSM) [9][11] close to PMRSM [5] has been presented with rather comprehensive solutions containing initial position estimation for nonhesitant starting via signal injection methods. As the PMRSM has notable saliency that is strongly inuenced by magnetic saturation, the vector-control has to account for it explicitly, both in the stator ux observers and in the control itself (in sensorless operation), especially when speed increases above the base speed and the stator ux is weakened due to battery dc voltage limitation. Sensorless direct torque and ux control with space vector modulation (DTFCSVM) for induction machines has been successfully proven, from very low speed, with robust dynamics, in a sliding-mode approach [12]. It has been applied also to reluctance synchronous motors [13]. The direct torque control (DTC) for PMSM has been reported recently [14], but not in the motion-sensorless SVM implementation, and not for ISA where the specications are much tougher than for industrial drives. The present paper develops a rather complete solution for sensorless DTFCSVM of PMRSM drive for integrated starter-alternator in mild hybrid electric vehicles, from zero speed to wide speed range, for motor/generator control. A stator ux reference with a quasioptimal ux versus torque functional is proposed. Stator ux observers in wide speed range employing combined voltage-current models for low speeds and only the voltage model for medium-high speeds in proportional-integral (PI) closed loops, are developed. A novel rotor speed and position observer with fusion strategy employs signal injection and only one D-module vector lter in stator referencefor low speeds, corroborated with a speed observer from the stator ux vector estimationfor medium-high speeds. The effectiveness of this solution has been conrmed by extensive experimental test results in wide speed range, including articial loading.

0885-8993/$20.00 2006 IEEE

712

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 3, MAY 2006

Fig. 1. Typical ISA requirements: (a) electric torque (T ) versus speed at crankshaft for motoring and (b) electric power (P ) and torque (T ) versus speed for generating, with various degrees of vehicle electrication.

II. TYPICAL INTEGRATED STARTER-ALTERNATOR SPECIFICATIONS The ISA has to perform for: 1) motoring: for vehicle self-starting (in torque control mode) and for auxiliaries driving during ICE shut-off (in speed control mode) and 2) generating: for vehicle regenerative braking and for battery recharge during exclusive ICE running. In essence, for both motoring and genversus speed erating regime, the electromagnetic torque (positive or negative) envelopes are given in Fig. 1. These envelopes have to be tempered further by the battery state of charge (accessibility) both in motoring and generating. The division of torque between ICE and PMRSM may also be decided by the optimum gas-mileage or lowest pollution criteria for the ICE. Further on, we will just retain the fact that: 1) the torque closed-loop control is mandatory and 2) an external power control-loop may be required for generating. As the battery power , this fact seems to be a ow can be easily measured good way to go.
Fig. 2. (a) PM-RSM cross section and (b) measured magnetization curves 9 (I ), 9 (I ).

A. PMRSM Model The PMRSM is a rather good saliency interior PM rotor synchronous machine (see Fig. 2) [7]. The rotor contains a multiple ux barrier structure with PMs placed on the bottom of ux barproduces a limited electromotive riers. The PM ux force (emf), of 150%200% at highest speed, to limit the inverter switches voltage rating (and costs). The PMs are placed in the -axis, while the -axis shows higher magnetic permeance . The PMRSM model in -rotor reference, considering the emf and inductances that sinusoidal vary with the rotor position, is given by

III. FLUX REFERENCE In what follows, the PMRSM is presented, with the novel stator ux referencer for the starter-alternator applications (for better performances in a high speed range).

(2)

BOLDEA et al.: DTFC-SVM MOTION-SENSORLESS CONTROL

713

Fig. 3. Two stator current pairs (I 9 or 9 .

;I

) for a given torque

and ux 9 = Fig. 5. PMRSM stator ux reference for starter-alternator.

reference ux functional, depending on the reference torque (see Fig. 4). On the other hand, for fast torque response in a machine which works mostly under heavy load and in ux weakening, the control goes to around zero stator ux at zero torque (point 0 in Fig. 4), when
Fig. 4. Flux-torque functional 9 (T ) from zero to base speed.

(4) When the maximum torque/ speed envelope is required, is applied and therefore and thus (5) Also for for (6)

where , and are the stator ux, current, and voltage vector, respectively, is the stator resistance; is the are the , inductances; is the electrical rotor speed; , pole-pair number; and means the conjugate vector operator. , The magnetic saturation is considered only in the -axis with constant, and cross-coupling saturation effects are neglected. B. Flux Reference is coordinated with In DTFC, the stator ux reference to produce the required (below maxthe torque reference imum) torque for a given speed. The problem is that for a given and torque there are two current pairs that ux can produce them

(3) The torque equation from (2) shows clearly that a specic , 0, when torque can be obtained with 0, for a given stator ux . An example is shown in the vector diagram in Fig. 3, based on the second equation from (2). , pair is better because the current amplitude is The 0). Thus, the smaller, and also the PM is less stressed ( is always in control should make sure that the ux vector the 4th quadrant for positive torque, or in the third quadrant for negative torque. One possibility to realize this fact in DTFC is for zero torque (point 0 to make the reference ux in Fig. 4), which guaranties zero stator current ( 0) for zero torque, and then the control starts from there to travel the

is a coefcient in the range of 0.950.97 for mowhere is the electrical rotor toring, and 1.031.07 for generating; base speed; is the stator voltage at base speed. This way, the and are provided functional (limiters) for. Whenever the required torque at a certain speed is smaller value, the control may than the available (maximum) value by maintaining the latter positive relax the

(7) For a given reference torque , the required -axis ux is computed, and then comes from curve. The novel funcon-line ux reference for the ux versus torque tional, valid for all situations, is introduced in Fig. 5. This is a quasioptimal choice. does not vary notably with stator The -axis inductance is current, and therefore the -axis magnetization curve used, that is very robust. The PI controller introduced in the ux referencer (see Fig. 5) is a very fast one and allows for solving

714

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 3, MAY 2006

Fig. 7. Speed and position observers with signal injection. (a) Observers with synchronous low-pass lter and with vector lter in D-module. (b) PLL used in (a).

Fig. 6. Stator ux observers with combined voltage-current models: (a) serial observer and (b) parallel observer.

an otherwise involved nonlinear equation. The ux-weakening zone is implicitly covered by the proposed ux referencer. The reference electric torque calculator from Fig. 5 reects the overall HEV control strategy and is not followed up here.

IV. SENSORLESS DTFC-SVM OF PM-RSM: THE OBSERVERS The comprehensive state observer includes the rotor speed and position observers and the stator ux observer. A. Stator Flux Observers For DTFC-SVM, the main estimations are for the stator ux and for the electromagnetic torque . The stator vector ux observers combine voltage-current models (see Fig. 6): the current model, including magnetic saturation, prevails at low speeds, while the voltage model prevails at medium-high speeds with a smooth transition between them, depending on speed. Two ux observers have been used for comparison. The serial observer [see Fig. 6(a)] contains the voltage model in stator reference serially connected with the current model in the rotor reference frame, using a PI compensator driven by the current estimation error. Both models are enclosed within the correction loop. The parallel observer [see Fig. 6(b)] contains the same models but connected in parallel and using the ux estimation error. This time, the PI correction loop encloses only the voltage model, while the current model is open loop operated.
Fig. 8. DTFCSVM motion-sensorless control of PMRSM.

Fig. 9. Experimental setup for sensorless DTFCSVM of PMRSM.

With accurate stator ux estimation and measured currents, the torque estimation is simply calculated in -stator reference as (8) is required in the current The rotor position estimation model within the ux observerbut only at low speed, below

BOLDEA et al.: DTFC-SVM MOTION-SENSORLESS CONTROL

715

Fig. 10. Test results at 5 rpm speed reversal without load: (a) from top to bottom: estimated speed (! ), measured speed (! ), estimated torque (T ), estimated ux components (9 ), and measured currents (I ); (b) estimated and measured rotor position ( ); c) estimated ux in  -plane.

50 rpm, because at medium-high speed the voltage model becomes dominant and the signal injection (which gives ) discontinuation does not produce a notable inuence on the ux observer. The rotor speed estimation in the whole speed range is required in the ux (and torque) referencer, and accuratelyin the speed controller. B. Speed and Position Observers With Signal Injection For low speed operation a speed observer using signal injection has been implemented as shown in Fig. 7. A rowith 1 V tating carrier-voltage 2 500 rad/s is injected in the stator superimand . posing on top of driving fundamental stator-voltage The resulting stator current , modulated by the position

dependent inductances with magnetic saliency, contains car. rier-current harmonics with rotor position information Usually, the negative rst-order sequence carrier harmonic with [9] is extracted from the measured stator current using various methods [15]. Two schemes have been tested: 1) observer with synchronous low-pass lter (LPF) in carrier reference [upper part of Fig. 7(a)] [10] and 2) novel observer with only one vector lter in D-module in stator reference as band-pass lter (BPF) with central frequency and phase-sign discrimination [lower part of Fig. 7(a)]. The mirror-phase vector control [11] uses two lters in D-module in sensorlees control. Both observers harmonic. A phase-locked loop (PLL) tracking select the

716

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 3, MAY 2006

observer shown in Fig. 7(b) extracts the estimation of rotor and position from the phase of harmonic in speed carrier reference, with PLL error . During the experiments it was noticed that both observers in Fig. 7(a) have similar behavior, with very robust estimation, almost independent of machine parameters. is correctly identied at standThe initial rotor position radians. still by both observers in the range of For full angle range, an uncertainty of radians appears, and this is discriminated by the magnet polarity detection using the second-order sequence carrier harmonics [9], [10]. C. Speed Estimation From Stator Flux For medium-high speed operation, above 50 rpm, the signal injection is disabled to reduce losses, and thus the associated speed and position observer from Fig. 7 is also inhibited. Now, additional speed estimation is required, with a smooth transition between the two of them. The proposed solution estimates the from the estimated stator ux vector rotor speed

(9) and are the estimated speed and position of in where -stator reference, which is given by the stator ux observer in is the estimated position of in -rotor reference Fig. 6; (see Fig. 3). The rst main component of the rotor speed in (9), i.e., is extracted together with from the phase by a PLL observer shown in Fig. 7(b). The second component, living only in transient regimes, is obtained from the electric torque expression that can be derived as

(10) The angle is extracted from a look-up table with (8) and from linear interpolation based on (10) using Fig. 6, with an average value for , because magnetic saturais calculated to nally tion is not so important. Then in (9). decide Note that, as the torque and ux response of the machine is faster than the speed response in automotive applications, the may alternatively be calculated as angle

Fig. 11. Test results at start-up to 10 rpm without load; from top to bottom: estimated speed (! ), measured speed (! ), estimated torque (T ), estimated ux components (9 ), and measured currents (I ).

(11) from the ux referencer in Fig. 5, provided that the ux and torque references do not experience step variations, or some ltering is applied. D. DTFCSVM Motion-Sensorless Control System Now that the proposed ux/torque referencers and the comprehensive state observers have been introduced, the proposed

sensorless DTFCSVM system for PMRSM is illustrated in Fig. 8 with the main components discussed above. Only the PI ux and torque regulators of DTFC-SVM have not been detailed so far. Their design is presented in [12], and is thus skipped here. In what follows, the digital implementation and the experimental results with the proposed DTFCSVM system will be tackled directly. V. EXPERIMENTAL RESULTS The experimental setup for the sensorless DTFCSVM of PMRSM (see Fig. 9) contains a belt transmission 10/13 kW (1500/3000 rpm) with an induction motor driven by an ABB ACS600 bidirectional inverter. The 140-Nm peak torque PMRSM is fed through a 48-Vdc, 350-A inverter from a 48-V, 55-Ah battery pack. The simulations (cut out from lack

BOLDEA et al.: DTFC-SVM MOTION-SENSORLESS CONTROL

717

Fig. 12. Test results at 750 rpm speed reversal without load; from top to bottom: estimated speed (! ) and measured speed (! ) (estimated from the stator ux and from signal injection at low speeds until 50 rpm), estimated torque (T ), estimated ux magnitude (9 )from serial and parallel observers.

Fig. 13. Articial loading with fast switching from motoring to generating: estimated speed (! ), measured speed (! ), estimated torque (T ), measured battery voltage (V ), measured battery current (I ).

of space) have been managed in Matlab/Simulink, and implemented on a dSpace 1103 control board. A series of experiments have been carried out to test the behavior of the proposed sensorless DTFCSVM of PMRSM. Two stator ux observers (see Fig. 6), and two rotor speed/position observers with signal injection (see Fig. 7) have been tested. The experimental test results at low and high speeds are shown. A fusion strategy (see Fig. 8) is proposed using the speed/position observer with signal injection for low-speed, and the speed estimation from stator ux for medium-high speed. Fig. 10 shows the no-load speed reversal at very low speed: from 5 rpm to 5 rpm. The ux hodograph is almost circular [see Fig. 10(c)], the position estimation is very good [see Fig. 10(b)] and the speed estimation is acceptable. The use of

SVM in DTFC has made a smooth operation possible at 5 rpm. Fig. 11 shows the no-load start from zero to 10 rpm, again with good results. The speed reversal at high speed from 750 rpm to 750 rpm, which requires the transition of the speed observer from signal injection at low-speed to estimation from stator ux at highspeed, is illustrated in Fig. 12 with good results. The switching from motoring to generatingas in articial loading [16], by pulsating the speed reference, is shown in Fig. 13, where the battery current changes sign with fast response. This articial loading has allowed for reaching 35 Nm peak torque with substantial battery charging (negative) current. VI. CONCLUSION The proposed sensorless DTFCSVM of PMRSM for HEV includes the following contributions.

718

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 3, MAY 2006

A quasioptimal ux referencer to optimize the energy consumption in wide speed range, with a ux versus torque functional leading to unique ux and torque reference values. A stator ux observer that uses a combined current-voltage models with PI compensator for low speed below 50 rpm, and only the voltage model in PI closed loop for mediumhigh speed. A speed and position observer with fusion strategy that employs signal injection and only one D-module vector lter in stator reference for low speed below 50 rpm, corroborated with a speed observer from stator ux vector estimation for medium-high speed. Extensive experimental test results on a laboratory prototype focused on very low speed to high speed sensorless operation for motoring and generating, including articial loading, prove the effectiveness of the proposed solution. Further investigations on online initial position estimation in full range, generator, and motor heavy loading over the entire speed range are still to be added to take full advantage of the sensorless DTFCSVM robust control capabilities.

[12] C. Lascu, I. Boldea, and F. Blaabjerg, Direct torque control of sensorless induction motor drives: a sliding-mode approach, IEEE Trans. Ind. Appl., vol. 40, no. 2, pp. 582590, Mar./Apr. 2004. [13] I. Boldea, L. Janosi, and F. Blaabjerg, A modied direct torque control (DTC) of reluctance synchronous motor sensorless drive, Electr. Machines Power Syst., vol. 28, no. 2, pp. 115128, Feb. 2000. [14] M. F. Rahman, M. E. Haque, L. Tang, and L. Zhong, Problems associated with the direct torque control of an interior permanent-magnet synchronous motor drive and their remedies, IEEE Trans. Ind. Electron., vol. 51, no. 4, pp. 799809, Jul./Aug. 2004. [15] G. D. Andreescu and I. Boldea, Integrated sensors of rotor position and speed based on signal injection for IPM-synchronous motor drives, in Proc. 8th IEEE Int. Conf. Intell. Eng. Syst. (INES04), Cluj-Napoca, Romania, Sep. 2004, pp. 371375. [16] L. Tutelea, I. Boldea, E. Ritchie, P. Sandholdt, and F. Blaabjerg, Thermal testing for inverter-fed induction machines using mixed frequency method, in Proc. Int. Conf. Elect. Machines (ICEM98), Istanbul, Turkey, Aug. 1998, vol. 1, pp. 248253.

REFERENCES
[1] J. M. Miller, V. R. Stefanovic, and E. Levi, Prognosis for integrated starter alternator systems in automotive applications, in Proc. 10th Int. Power Electron. Motion Contr. Conf. (EPE-PEMC02), Dubrovnik, Croatia, Sep. 2002, pp. 115. [2] A. Vagati, A. Fratta, P. Guglielmi, G. Franchi, and F. Villata, Comparison of ac motor based drives for electric vehicles, in Proc. Power Conv. Intell. Motion Conf. (PCIM99), Nuremberg, Germany, Jun. 1999, vol. I, pp. 173180. [3] T. Teratani, K. Kuramochi, H. Nakao, T. Tachibana, K. Yagi, and S. Abou, Development of Toyota mild hybrid system (THS-M) with 42 V power-net, in Proc. IEEE Int. Elect. Machines Drives Conf. (IEMDC03), Madison, WI, Jun. 2003, vol. 1, pp. 310. [4] A. Lange, W.-R. Canders, F. Laube, and H. Mosebach, Comparisons of different drive systems for a 75 kW electrical vehicle drive, in Proc. Int. Conf. Elect. Machines (ICEM00), Espoo, Finland, Aug. 2000, vol. 3, pp. 13081312. [5] P. Guglielmi, M. Pastorelli, G. Pellegrino, and A. Vagati, Positionsensorless control of permanent-magnet-assisted synchronous reluctance motor, IEEE Trans. Ind. Appl., vol. 40, no. 2, pp. 615622, Mar./Apr. 2004. [6] E. C. Lovelace, T. Keim, J. H. Lang, D. D. Wentzloff, T. M. Jahns, J. Way, and P. J. McCleer, Design and experimental verication of a direct-drive interior PM synchronous machine using a saturable lumpedparameter model, in Proc. 37th IEEE-IAS02 Annu. Meeting, Pittsburgh, PA, Oct. 2002, vol. 4, pp. 24862492. [7] I. Boldea, L. Tutelea, and C. I. Pitic, PM-assisted reluctance synchronous motor/ generator (PM-RSM) for mild hybrid vehicles: electromagnetic design, IEEE Trans. Ind. Appl., vol. 40, no. 2, pp. 492498, Mar./Apr. 2004. [8] B. A. Welchko, T. M. Jahns, W. L. Soong, and J. A. Nagashima, IPM synchronous machine drive response to symmetrical and asymmetrical short circuit faults, IEEE Trans. Energy Conv., vol. 18, no. 2, pp. 291298, Jun. 2003. [9] H. Kim, K.-K. Huh, R. D. Lorenz, and T. M. Jahns, A novel method for initial rotor position estimation for IPM synchronous machine drives, IEEE Trans. Ind. Appl., vol. 40, no. 5, pp. 13691378, Sep./Oct. 2004. [10] Y. Jeong, R. D. Lorenz, T. M. Jahns, and S. Sul, Initial rotor position estimation of an interior permanent magnet synchronous machine using carrier-frequency injection methods, in Proc. IEEE Int. Elect. Machines Drives Conf. (IEMDC03), Madison, WI, Jun. 2003, vol. 2, pp. 12181223. [11] S. Shinnaka, New mirror-phase vector control for sensorless drive of permanent-magnet synchronous motor with pole saliency, IEEE Trans. Ind. Appl., vol. 40, no. 2, pp. 599606, Mar./Apr. 2004.

Ion Boldea (M77SM81F96) received the B.S. and Ph.D. degrees in electrical engineering from the University Politehnica, Timisoara, Romania, in 1967 and 1973, respectively. He is currently a Full Professor at the University Politehnica Timisoara. He has worked and published extensively on linear and rotary electric machines and their power electronics control, with and without motion sensors. He co-authored Induction Machine Handbook (Orlando, FL: CRC, 2001), Linear Motion Electromagnetic Devices (London, U.K.: Taylor & Francis, 2001), Electric Drives (Boca Raton, FL: CRC, 2006), and The Electric Generators Handbook (Boca Raton, FL: CRC, 2006). He spent about ve years as a Visiting Scholar in the U.S. and U.K., and presented keynote addresses, intensive courses, and does consultant work in the U.S., Europe, and Asia. He is Associate Editor of the Electric Power Components and Systems Journal, and Director and Founder of the Internet-only, Journal of Electrical Engineering. Dr. Boldea is an active member of the Industrial Drives and Electric Machines Committee, IEEE Industry Applications Society and Co-chairman of the IEEE IAS Technically Sponsored OPTIM96, 98, 00, 02, 04, and 06 International Conferences.

Cristian Ilie Pitic was born in Hunedoara, Romania, on December 18, 1976. He received the Ph.D. degree in electrical engineering from the Department of Electrical Drives and Power Electronics, University Politehnica Timisoara, Timisoara, Romania, in 2005. In 2004, he was a Guest Researcher at the Institute of Energy Technology, Aalborg University, Aalborg East, Denmark. His research interests include design, testing, and control of permanent-magnet-assisted reluctance synchronous motors for electric and hybrid vehicles.

Cristian Lascu received the M.S. and Ph.D. degrees in electrical engineering from the University Politehnica Timisoara, Timisoara, Romania, in 1995 and 2002, respectively. In 1995, he joined the Department of Electrical Engineering, University Politehnica Timisoara where his research was focused on power electronics and high performance electrical drives. He was a Visiting Researcher at the Institute of Energy Technology, Aalborg University, Aalborg East, Denmark, in 1997 and 2005, and with the Department of Electrical Engineering, University of Nevada, Reno, from 1999 to 2000. From 2002 to 2004, he was with SIEI S.p.A., Italy, working on advanced power electronics and drives for electrical vehicles under a European Marie Curie Fellowship. Dr. Lascu received the IEEE-IAS Prize Paper Award in 1998.

BOLDEA et al.: DTFC-SVM MOTION-SENSORLESS CONTROL

719

Gheorghe-Daniel Andreescu (M03SM05) received the M.S. degree in applied electronics and the Ph.D. degree in automatic control systems from the University Politehnica Timisoara, Timisoara, Romania, in 1977 and 1999, respectively. From 1977 to 1984, he was an R&D Team Leader in applied electronics and industrial control with the Electrotimis Company, Timisoara. Since 1984, he has been a member of the Department of Automation and Applied Informatics, University Politehnica Timisoara, where he is currently a Full Professor. Since 1993, he has also been a Researcher with TIMTEH Electronics Company, Timisoara, where from 1994 to 2002, he specialized in automatic testing for avionics, e.g., at British Airways Avionic Engineering (BAAE), Cardiff, U.K. He is the author or coauthor of over 50 papers in journals and international conferences, two books, and ve textbooks. His major book is Estimators in Control of Electric Drives. Applications to PMSMs (Timisoara, Romania: EOU, 1999). His research interests are mainly in the areas of advanced control of ac drivessensorless control, electric/ hybrid vehicles, observers, sliding-mode control, mechatronic systemsrobotics, automatic testing for avionics, microcontroller and VHDL applications, real time implementations, monitoring and control of distributed systems. Dr. Andreescu is a member of the Technical Committee TC4.2 on Mechatronic Systems of the International Federation of Automatic Control (IFAC), the Romanian Society of Control Engineering and Technical Informatics, and the Romania Association of Robotics.

Lucian Tutelea was born in Alba, Romania. He received the B.E. and Ph.D. degrees in electrical engineering from the University Politehnica Timisoara, Timisoara, Romania, in 1989 and 1997, respectively. He is currently a Lecturer in the Department of Electric Drives and Power Electronics, University Politehnica Timisoara. His research interests include design and control of electrical machines and drives.

Frede Blaabjerg (S86M88SM97F03) was born in Erslev, Denmark, on May 6, 1963. He received the M.Sc.EE. and Ph.D. degrees from Aalborg University, Aalborg East, Denmark, in 1987 and 1995, respectively. He was with ABB-Scandia, Randers, Denmark, from 1987 to 1988. He is currently Full Professor in power electronics and drives and Dean of the Faculty of Engineering and Science, Aalborg University. In 2000, he was a Visiting Professor with the University of Padova, Padova, Italy, as well as part-time Programme Research Leader in wind turbines at Research Center Risoe, Risoe, Italy. In 2002, he was a Visiting Professor at the Curtin University of Technology, Perth, Australia. He is involved in more than 15 research projects within the industry including the Danfoss Professor Programme in Power Electronics and Drives. He is the author or co-author of more than 350 publications in his research elds including the book Control in Power Electronics (New York: Academic, 2002). He is an Associate Editor for the Journal of Power Electronics and for Elteknik. His research areas are in power electronics, static power converters, ac drives, switched reluctance drives, modeling, characterization of power semiconductor devices and simulation, power quality, wind turbines, and green power inverters. Dr. Blaabjerg received the 1995 Angelos Award for his contribution in modulation technique and control of electric drives, the Annual Teacher Prize from Aalborg University in 1995, the Outstanding Young Power Electronics Engineer Award from the IEEE Power Electronics Society in 1998, ve IEEE Prize Paper Awards during the last six years, the C. Y. OConnor Fellowship from Perth, Australia in 2002, the Statoil-Prize for his contributions in Power Electronics in 2003, and the Grundfos Prize in acknowledgment of his international scientic research in power electronics in 2004. He has been an Associate Editor of the IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, and currently is Editor-in-Chief of the IEEE TRANSACTIONS ON POWER ELECTRONICS. He is a member of the European Power Electronics and Drives Association, the IEEE Industry Applications Society Industrial Drives Committee, the Industry Power Converter Committee and the Power Electronics Devices and Components Committee, IEEE Industry Application Society. He has served as member of the Danish Technical Research Council in Denmark from 1997 to 2003 and from 2001 to 2003 he was Chairman. He has also been Chairman of the Danish Small Satellite Programme and the Center Contract Committee. He became a member of the Danish Academy of Technical Science in 2001 and in 2003 he became a member of the Academic Council. From 2002 to 2003, he was a member of the Board of the Danish Research Councils. In 2004, he became Chairman of the Programme Committee on Energy and Environment.

Per Sandholdt (S92M94) was born in Kolding, Denmark, on May 3, 1967. He received the M.Sc.EE. and Ph.D. degrees from Aalborg University, Aalborg East, Denmark, in 1993 and 2002, respectively. From 1998 to 2001, he was Assistant Professor at Aalborg University. In 2001, he joined Sauer-Danfoss, Nordborg, Denmark, where he is currently working with electric motor drives in the cooperate-R&D Department.