Electric Vehicle Traction based on a PM Assisted

Electric Vehicle Traction based on a PM Assisted Synchronous Reluctance Motor Nicola Bianchi, Emanuele Fornasiero, Enrico Carraro, Silverio Bolognani, Mos`e Castiello Department of Industrial Engineering, University of Padova, via Gradenigo 6 A, Padova, Italy

Abstract—It is recently demonstrated that the synchronous reluctance motor is well suited for electric as well as for hybrid electric vehicles. Of course, a proper rotor design is necessary, since the main torque is due to the rotor anisotropy, that is, the permeance difference between the direct- and the quadrature-axis. Then, the flux-barrier ends are placed so as to reduce the torque ripple due to the slot harmonics. The torque ripple is also reduced adopting the ”Machaon” configuration, which includes rotor asymmetries. The impact of the rotor design on the motor performance is presented deeply, showing several experimental test results carried out on synchronous reluctance motors with different rotor geometries. The impact of the assistance of Ferrite permanent magnet (PM) inset in the fluxbarriers is also shown: highlighting the main benefits of the PM assistance to the synchronous reluctance motor capabilities.

I. I NTRODUCTION Interior permanent magnet (IPM) motors with rare-earth magnets are mainly used for electric vehicles (EVs) and hybrid EVs (HEVs). They offer high capabilities, particularly high torque density and constant power operation in a wide speed range. However, due to high cost of rare-earth magnets and to limited supply, IPM motors are becoming too expensive. Therefore the synchronous reluctance (REL) machine is becoming of great interest in the recent years. If properly designed, it represents a valid alternative for EVs and HEVs for its simple and rugged construction and for hazard-free operations [1], [2]. The REL motor as well as the Ferrite PM assisted REL (PMAREL) motor are becoming competitors of both surface–mounted PM machines and induction machines not only in automotive but also in many other applications [3].

Flux barriers Iron bridges to sustain the structure

Magnets assisting reluctance motor

Rotor lamination

along the q-axis the flux lines are obstructed

along the d-axis the flux lines are not obstructed

Fig. 1: Synchronous reluctance rotor

A four–pole REL is sketched in Fig. 1. It refers to a rotor with three flux barriers per pole. A rotor configuration with several flux barriers per pole allows to achieve a high rotor saliency, that is, a high average torque. In designing such a REL motor, the main attention has to be given (i) to maximize the average torque, (ii) to minimize the torque ripple and (iii) to maximize the motor efficiency in an extended speed range. The rotor geometry has a high influence on the machine performance, in terms of both average torque and ripple. An optimization is often required to the aim of determining a rotor geometry achieving a high and smooth torque [4]. The synchronous PMAREL motor is achieved when PMs are inset within the flux barriers [5], [6]. The inset of PMs within the flux barriers tends to increase not only the average torque but the power factor (PF), which is commonly quite low in a REL motor, so that the rquired Volt-Amps power rating is reduced. In fact, the PM flux saturates the iron bridges, reducing the magnetizing stator current, and

tends to rotate the flux linkage vector out of phase of 90 degrees with respect the current vector.

Fig. 3: Average torque versus current angle of non skewed PMAREL motor. Currents used in the tests are 5 A, 10 A, 15 A and 20 A (experimental results)

Torque [Nm]

Fig. 2: Average torque versus current angle of non skewed REL motor. Currents used in the tests are 5 A, 10 A, 15 A and 20 A (experimental results)

II. T ORQUE VERSUS CURRENT VECTOR ANGLE

12 11 0

90

180 θ [degrees]

270

360

270

360

m

(a) non skewed rotor Torque [Nm]

A prototype (main data are in Appendix) design for such an application is tested experimentally. Figs. 2 shows the measured torque versus the current angle for the REL machine, according to different current amplitudes. It refers to the motor prototype without skewing. At the same current, the torque is slightly lower for the motor with skewed rotor, expecially at higher current. At 20 A, the torque difference is about 6%. The current angle corresponding to the MTPA trajectory, instead, remains quite constant with the current amplitude, being quite unchanged for the two motor prototypes. To highlight the assistance of PMs, Fig. 3 shows the torque versus the current angle of the PMAREL machine, without skewing, according to different current amplitudes. Ferrite PMs are used in the prototype, characterized by a remanence Bres = 0.4 T (at 20o C). Torque of the motor with step–skewed rotor is about 5% lower. As expected, comparing the torque exhibited by the REL motors (Fig. 2) and by PMAREL motors (Fig. 3) it is verified that there is a higher average torque when the PMs assist the motors. According to a current of 10 A, the average torque increases from 11.3 Nm to 12.1 Nm. According to a current of 15 A, the average torque increases from 17.7 Nm to 19.0 Nm. In addition, the current angles corresponding to the MTPA trajectory are slightly lower for the PMAREL motors. In fact, the current vector moves in the direction of the added PM flux linkage vector.

13

13 12 11 0

90

180 θ [degrees] m

(b) skewed rotor

Fig. 4: REL motor: measured torque versus mechanical position I = 10 A

A. Skewing effect on torque ripple Due to the interaction between the spatial harmonics of magneto-motive force (MMF) and the rotor geometry, the REL machine is characterised by a high torque ripple [7]–[10]. Fig. 4 shows the measured torque versus the mechanical position for the REL motor, when it is supplied with a stator current I = 10 A, without and with rotor skewing. The torque ripple decreases from about 17% to about 9% of the average torque. Similar results are measured in the PMAREL. When PMs are used a step–skewing is adopted: the rotor is split in three parts, each of them is skewed with respect to the others. Fig. 5 shows the measured torque behaviour of the PMAREL motors, according to the same stator currents I = 10 A.

Torque [Nm]

TABLE I: REL motor: steady state operations

13 12

90

180 θm [degrees]

270

250

2 4 6 8 10 12

2.26 3.35 4.4 5.39 6.42 7.43

0.64 0.72 0.78 0.81 0.83 0.85

52 58 58 57 54 52

2.00 3.07 4.06 5.03 6.03 6.99

0.71 0.78 0.84 0.87 0.88 0.90

68 66 64 62 59 57

500

2 4 6 8 10 12

2.25 3.34 4.38 5.38 6.38 7.44

0.70 0.75 0.79 0.82 0.83 0.84

61 69 71 70 69 67

2.00 3.07 4.06 5.03 6.03 6.99

0.77 0.81 0.86 0.87 0.88 0.89

79 77 77 75 73 71

360

Torque [Nm]

(a) non skewed rotor 13 12 11 0

90

180 θm [degrees]

270

360

PMAREL motor I PF∗ η (A) (%)

I (A)

11 0

REL motor PF∗ η (%)

T (N m)

n (rpm)

(b) skewed rotor

Fig. 5: PMAREL motor: measured torque versus mechanical position I = 10 A

As expected [4], the torque ripple is not completly reduced by means of the rotor skewing. In fact, the same stator current yield different d– and q–axis components along the rotor axial length when the rotor is skewed. III. S TEADY- STATE O PERATIONS The tests of the steady-state performance have been carried out feeding the motors by different currents I in order to produce a given output torque, at given rotation speeds. Table I reports some measurements at steady-state operations, for torque values in the range between 2 and 12 Nm. The d– and q–axis currents are selected so as to operate along the MTPA trajectory. Even though the PM is a low–energy (Ferrite) magnet, it is worth noticing that the insertion of the PMs allows a lower stator current to be required. At low speed, the motor efficiency results to be quite low, but enough to verify the impact of the PM on the losses reduction, for given torque. In the two tables the power factor (PF) is reported. The PF reported in the table has to be considered only as an index, whose value is close to the PF (for this reason a star is added as superscript, P F ∗ ). Anyway, it is possible to observe an increase of this index in the PMAREL motor. A higher increase

Fig. 6: REL motor: torque and power versus speed. Experimental versus simulated results.

is expected when comparing the fundamental harmonic waveforms. IV. F LUX –W EAKENING P ERFORMANCE The motors have been tested also for various rotating speeds to determine their Flux–Weakening (FW) performance [11], [12]. Fig. 6 shows the torque and power versus speed for the REL machine. Solid line refers to the simulated curve, while circles refer to the measurements. Fig. 7 shows the torque and power versus speed for the PMAREL machine. There is a good agreement between the simulated and the measured operating points, even if at higher speeds the predicted torque is quite higher than the measured torque for the PMAREL motor. V. A SYMMETRIC ROTOR F LUX -BARRIERS When the flux barrier geometry is different in adjacent poles, a sort of compensation of the torque

(a) Symmetric rotor A-type

Fig. 7: PMAREL motor: torque and power versus speed. Experimental versus simulated results. (b) Symmetric rotor B-type

(c) Asymmetric rotor

Fig. 8: Geometries of symmetric and asymmetric (”Machaon“) rotor

torque (Nm)

harmonics is achieved [13]–[15]. The resulting motor is referred to as ”Machaon“ motor (the name of a butterfly with two large and two small wings), since the flux barriers of the adjacent poles are large and small alternatively. Fig. 8 shows three different rotors: two rotor with symmetric flux barrier geometry (A–type and B– type) and a rotor with asymmetric rotor, that is, a machaon–type rotor. In this third rotor, the PMs remain in the same position, and only the lateral part of the flux barriers and the end angles are modified. Fig. 9 shows the torque versus the rotor position θm of the REL motor. The motors are fed with the same current, amplitude and phase angle. The motors with symmetric A–type and B–type rotors exhibit a torque ripple, whose the harmonic of 18th order is well recognized (i.e., with three periods each 30 mechanical degrees). This is expected since this is the slot harmonic. Let’s note that these two A–type and B–type motors exhibit such a torque harmonic of 18th order with almost the same amplitude but out of phase of about 180 degrees. When these two geometries are combined together, and the machaon–type rotor is achieved, there is a sort of compensation of the harmonic of 18th order. The torque ripple of the machaon–type motor exhibits no harmonic of 18th order, so that the torque harmonic of 36th order is more evident (i.e., with six periods each 30 mechanical degrees). With this geometry the torque ripple has been found to be reduce to two third, for both REL and PMAREL motor. In addition, the skewing angle can

15 14 13 12 11 10 9

A

0

5

10

15 20 rotor position (deg)

B

25

30

Fig. 9: REL motor: torque behaviour with symmetric and asymmetric rotor (I=10 A, αie =64 deg)

be reduced. On the other hands, the average torque remains almost the same, as highlighted in Fig. 9 by the thin solid line. A PPENDIX Test bench description The typical stator of a induction motor is used for the synchronous REL motors under test. The slot number is 36 and the back iron is designed

according to four poles. Outer and inner diameter are De =200 mm and Di =125 mm, respectively, and the stack length is Lstk =40 mm. Fig. 10 shows a picture of the test bench used for these measurements. The motor under test is on the left hand side. The master machine is on the right hand side. The torquemeter can be seen between the two machines.

Fig. 10: Test bench for motor test. Several experimental tests are reported according to a synchronous reluctance motor and a PM assisted reluctance. Introducing PMs in the rotor, (i) the torque increases of about 10%, mainly in FW operations where a wide constant power–speed range is achieved, and (ii) the power factor improves in the whole operating region. With a rotor skewing, (i) the average torque slightly decreases, but (ii) the torque ripple decreases down to about one third. With an asymmetrical rotor (Machaon geometry): (i) the average torque remains the same, and (ii) the torque ripple decreases. Further details will be given in the full paper. The REL motor seems to be a good competitor for electric vehicles, exhibiting high torque density, properly low torque ripple, and a high overload capability. Its construction is robust and it is free from rare–earth magnets, which makes it very actractive from the cost point of view. R EFERENCES [1] K. M. Rahman, B. Fahimi, G. Suresh, A. V. Rajarathnam, and M. Ehsani, “Advantages of switched reluctance motor applications to ev and hev: Design and control issues,” IEEE Trans. on Industry Applications, vol. 36, no. 1, Jan/Feb 2000. [2] D. A. Staton, T. J. E. Miller, and S. E. Wood, “Maximising the saliency ratio of the synchronous reluctance motor,” IEEE Trans. on Industry Applications, vol. 140, no. 4, July 1993.

[3] M. Barcaro and N. Bianchi. “Interior PM Machines using Ferrite to Substitute Rare–Earth Surface PM Machines.” Conf. Rec. of Int. Conf. of Electr. Machines, ICEM, Marsille (F), pp. 1–7, June 2012. [4] A. Vagati, M. Pastorelli, G. Franceschini, and S.C. Petrache. ”Design of low-torque-ripple synchronous reluctance motors.” IEEE Trans. on Industry Application, IA-34(4):758–765, July– Aug. 1998. [5] A. Fratta, A. Vagati, and F. Villata. ”Permanent magnet assisted synchronous reluctance drive for constant-power application: Drive power limit.” In Proc. of Intelligent Motion European Conference, PCIM, pages 196–203, April Nurnberg, Germany, 1992. [6] W.H. Kim, K.S. Kim, S.J. Kim, D.W. Kang, S.C. Go, Y.D. Chun, J. Lee, “Optimal PM Design of PMA-SynRM for Wide ConstantPower Operation and Torque Ripple Reduction,” IEEE Transactions on Magnetics, vol. 45, no. 10, pp. 4660–4663, Oct. 2009. [7] A. Fratta, G. P. Troglia, A. Vagati, and F. Villata, “Evaluation of torque ripple in high performance synchronous reluctance machines,” in Industry Applications Society Annual Meeting, 1993., Conference Record of the 1993 IEEE, Toronto, Ont., Oct. 1993, pp. 163–170. [8] T. M. Jahns and W. L. Soong, “Pulsating torque minimization techniques for permanent magnet AC motor drives-a review,” IEEE Transactions on Industrial Electronics, vol. 43, no. 2, pp. 321–330, Apr. 1996. [9] J.M. Park, S.I. Kim, J.P. Hong, J.H. Lee, “Rotor Design on Torque Ripple Reduction for a Synchronous Reluctance Motor With Concentrated Winding Using Response Surface Methodology,” IEEE Transactions on Magnetics, vol. 42, no. 10, pp. 3479–3481, Oct. 2006. [10] S.-H. Han, T. Jahns, W. Soong, M. Guven, and M. Illindala, “Torque ripple reduction in interior permanent magnet synchronous machines using stators with odd number of slots per pole pair,” Energy Conversion, IEEE Transactions on, vol. 25, no. 1, pp. 118–127, 2010. [11] A. Fratta, A. Vagati, and F. Villata, “Permanent magnet assisted synchronous reluctance drive for constant-power application: Drive power limit,” in Proc. of Intelligent Motion European Conference, PCIM, April Nurnberg, Germany, 1992, pp. 196– 203. [12] W.-H. Kim, K.-S. Kim, S.-J. Kim, D.-W. Kang, S.-C. Go, Y.-D. Chun, and J. Lee, “Optimal pm design of pma-synrm for wide constant-power operation and torque ripple reduction,” Magnetics, IEEE Transactions on, vol. 45, no. 10, pp. 4660– 4663, 2009. [13] N. Bianchi, S. Bolognani, D. Bon, and M. Dai Pr`e, “Rotor flux-barrier design for torque ripple reduction in synchronous reluctance and pm-assisted synchronous reluctance motors,” IEEE Transactions on Industry Applications, vol. 45, no. 3, pp. 921– 928, 2009. [14] M. Barcaro and N. Bianchi, “Torque Ripple Reduction in Fractional-Slot Interior PM Machines Optimizing the FluxBarrier Geometries,” in International Conference on Electrical Machines (ICEM), 2012, sept. 2012. [15] L. Alberti, M. Barcaro, and N. Bianchi “Design of a Low Torque Ripple Fractional-slot Interior Permanent Magnet Motor.” in Conf. Rec. of the 2012 IEEE Energy Conversion Conference and Exposition, ECCE, Raleigh NC, USA, vol. 1, pp. 1–8, 2012.