Research Article  Open Access
Investigation of a New FluxModulated Permanent Magnet Brushless Motor for EVs
Abstract
This paper presents a fluxmodulated direct drive (FMDD) motor. The key is to integrate the magnetic gear with the PM motor while removing the gear innerrotor. Hence, the proposed FMDD motor can achieve the lowspeed hightorque output and highspeed compact design requirements as well as hightorque density with a simple structure. The output power equation is analytically derived. By using finite element analysis (FEA), the static characteristics of the proposed motor are obtained. Based on these characteristics, the system mathematical model can be established. Hence, the evaluation of system performances is conducted by computer simulation using the Matlab/Simulink. A prototype is designed and built for experimentation. Experimental results are given to verify the theoretical analysis and simulation.
1. Introduction
To improve the fuel economy and reduce the emissions in the transportation areas, electric vehicles (EVs) and hybrid electric vehicles (HEVs) are the most viable solution [1]. As the key parts of EVs and HEVs, electric motors can be classified into two types. One is the concentrate motor driving [2]; the wheels are driven by motor through planetary gears and gearboxes indirectly. However, it causes the maintenance, transmission loss, and acoustic noise. The other is the inwheel motor driving; especially the PM inwheel motors have been paid much attention, which have the advantages of simple structure and high efficiency. There are two types of PM inwheel motors; one is the outerrotor topology without gear; another is the innerrotor one in which a planetary gear is employed. The former can provide the lowspeed operation directly, but it causes big size and heavy weight. On the other hand, the latter has the advantages of reduced size and weight, but the planetary gear has the defects such as lubrication, transmission loss, and acoustic noise [3].
With the advent of magnetic gears, magnetic transmission systems have been developed quickly [4]. The magnetic gear has some distinct advantages when compared to mechanical gears: no mechanical fatigue; no lubrication; overload protected; no contact losses; no transmission contact acoustic noise; high efficiency (only a little core loss and bearing loss); and high torque per volume ratio (ten times the standard motors) [5]. When such a magnetic gear is coupled with a conventional PM motor, the overall torque density can be significantly improved. Based on this combination, a magneticgeared PM motor is proposed to be used as a direct drive motor [6]. Figure 1(a) shows the configuration of this magneticgeared PM motor with three airgaps, which consists of four parts: the stator, modulation ring, inner rotor, and outerrotor. The modulation ring is used to modulate the airgap field space harmonics. The PMs are buried in the iron core of both rotors and magnetized in alienpolarity. However, the complicated structure causes manufacturing difficulty and instability. Furthermore, its power density is limited by its high flux leakage and iron loss. In [7], an improved topology with two airgaps is proposed, in which the highspeed innerrotor is omitted, so the structure is simple. But the outerrotor is the same as the three airgap topology. In [8], another fractionalslot fluxmodulated PM motor with two airgaps is developed. The rule for comparing the power density of electric motors is proposed and its cogging torque is very small. It also confirms that the magneticgeared PM motor is a better choice than the conventional PM motor for lowspeed drives. However, the same requirement of two or three airgaps as that in [5–9] will lead to the same problems. Furthermore, experimental results are rarely given to verify the design, analysis, and control of magneticgeared PM motors.
(a)
(b)
The major contribution of this paper is to propose and implement a fluxmodulated direct drive (FMDD) motor, having a simple structure while incorporating the advantages of PM machines and magnetic gears. This paper will focus on the design, analysis, control, and experimental verification of the proposed motor. The motor configuration and operating principles will be described in Section 2. The design criteria of the motor are presented in Section 3. Section 4 will be devoted to the electromagnetic analysis. The twodimensional (2D) FEA is used to analyze the static performance of the motor. In Section 5, based on Matlab/Simulink, simulation of the whole drive system will be discussed. The implementation of testbed and experimental results will also be given for verification. Finally, conclusion will be drawn in Section 6.
2. Motor Configurations and Operating Principle
2.1. Motor Topology
As shown in Figure 1(b), it is the proposed fluxmodulated direct drive motor for lowspeed drives. The coaxial magnetic gear is integrated into a PM machine; this gear is composed of three parts which are stator, outerrotor, and stationary ring between them. It should be noted that the stationary ring can be separated into two parts: one is the flux modulation block which is made from laminated iron core, and it provides the path for magnetic field, and the remaining part is filled with epoxy which is nonferromagnetic. The highspeed rotating field of the inner armature windings is modulated into the lowspeed rotating field of the PM outerrotor. Hence, the outerrotor can provide a hightorque output while operating at a low speed.
All the PMs are embedded in the outerrotor along the circumferential periphery and magnetized in the same direction. PM flux flowing in the iron core is equivalent with that magnetized in the alien direction, thus creating a distributed magnetic field with the same number of pole pairs. Compared with the motor in Figure 1(a), the PM materials of the proposed motor can be saved, and the structure is simplified. Hence, cost can be reduced and reliability can be improved as well.
2.2. Operating Principle
The proposed motor operates on principle of coaxial magnetic gear. According to [4], the pole pairs of flux density space harmonics distribution produced by the PM outerrotor can be deduced as shown in (1), and the speed of the flux density space harmonics is given by (2): where , ,is the pole pairs of the PM outerrotor, is the number of the magnetic field modulation blocks of stationary ring, and is angular velocity of the PM outerrotor.
To transmit a steady torque, the pole pairs of stator magnetic field must be equal to the pole pairs of the space harmonic flux density distribution . To utilize the maximum harmonic field, , . So,, and are governed by
The speed ratio is given by where is the speed of rotating field in stator and the negative sign indicates that the direction of rotating field in stator is opposite to that of the PM outerrotor.
For the proposed machine, , , , and the speed ratio.
3. Design Criteria
3.1. Sizing Equation
To reduce THD of back EMF, fractional slot shorted pitch distribution windings are used in [10, 11]. Assuming the stator winding current is controlled to be sinusoidal and synchronized with the back EMF and leakage inductance and winding resistance are ignored, based on [12], the stator sizing equation of the proposed motor can be expressed as whereis outside diameter of stator,is active axial length,is output power,is efficiency,is armature winding factor,is the fundamental amplitude of airgap flux density,_{ }andis electric loading.
Whenis determined, it can be observed that the main dimensions can be designed based on motor speed and electromagnetic load.
3.2. Design Flow
The design flow of the proposed motor is given below.(1)According to the technical requirements, the rated power, speed, torque, and voltage of the motor can be deduced.(2)Based on (5), stator diameter, stator pole pairs, rotor pole pairs, stack length, and stator winding turns can be initialized.(3)Determine the solution domain and the boundary conditions for finite element analysis, generate meshes, and apply the FEM for magnetic field analysis.(4)Evaluate the magnetic field distributions, airgap flux linkages, and inductances. Hence, the noload EMF and torque performances can be simulated.(5)Modify the size of motor and winding turns, and repeat steps to until the expected performances can be attained.
4. Static Performance Analysis
The 2D finite element method (FEA) is employed to analyze the static performance of the proposed motor. The specifications of the proposed motor are listed in Table 1. Figures 2(a) and 2(b) compare the magnetic field distributions at noload and fullload. It can be observed that most of the flux produced by the PMs embedded in the outerrotor passes through the modulation blocks of the stationary ring. In Figure 3 are the radial magnetic field waveforms in both the inner airgap and the outer airgap at noload. It can be noted that the largest space harmonic is changed from 23 pole pairs in the outer airgap to 4 pole pairs in the inner airgap, which verifies the modulation effect of the stationary ring.

(a)
(b)
(a)
(b)
(c)
(d)
Cogging torque is generally a major problem in PM motors [13–16]. In the proposed motor, various techniques such as fractional slot, optimized pole arc width, and optimal number of modulation blocks are adopted to reduce the cogging torque. Figure 4 shows the cogging torque waveform calculated by the FEA. The magnitude of the cogging torque of the proposed motor is very small, only 0.05 Nm, which is due to the large value of the smallest common multiple between stator pole pairs and the rotor pole pairs.
5. Simulation and Experimental Verification
5.1. System Configuration and Control Strategy
A 500 W target motor is built to verify the validity of the proposed motor as shown in Figure 5. In order to verify the performance of the proposed motor, an integrated power stage is used to drive the prototype, which is composed of a threephase diode bridge rectifier and an inverter controlled by a DSP. The current inputs, encoder circuit and protection, are also included in the controller circuit board. The system configuration schematic is shown in Figure 6. The drive system includes the prototype motor, a SEMIKRON integrated fullbridge inverter, and a DSPbased controller. An incremental encoder with 1024 pulses/revolution is employed in this setup. A DC dynamometer is a variable load. The output torque of the proposed machine is measured by a HBM torque transducer, and all data are acquired via a computer.
(a)
(b)
(c)
(d)
Because the back EMFs are sinusoidal, the turnon angle can be selected to ensure that phase current has fixed phase difference with the back EMF. Based on the back EMF profiles (, , ) shown in Figure 7(a), the control logic for all power switches of the inverter (, , , , , ) can be determined as shown in Figure 7(b).
(a) Back EMF profiles
(b) Power switches of the inverter
5.2. Simulation and Experimental Results
Figure 8 shows the measured opencircuit phasetoneutral back EMF waveform at the rated speed of 400 rpm. Figure 9 shows the measured back EMF waveform of phase A for the prototype together with the predicted waveform obtained by FEA. It can be seen that the measured back EMF waveform well agrees with the simulation one and both are very sinusoidal. The harmonic analysis of the back EMF of phase A deduces that the THD is 2.81%. The corresponding histogram is depicted in Figure 10. Also, from Figure 8, it can be observed that the experimental back EMF frequency is 153 Hz. It indicates that the rotational speed of the magnetic field in the inner airgap is 2299 rpm (namely, 5.749 times that of outerrotor speed) which agrees with theoretical value of 5.75. The result confirms the validity of the design. Meanwhile, there is a slight difference between the amplitudes of these two waveforms. The FEApredicted back EMF waveform for the proposed motor has the amplitude 5.88% higher than that of the measured value. This difference is caused by two reasons. Firstly, there are inevitable errors caused by manufacturing process. Secondly, the 2D FEA is definitely less accurate than the 3D FEA.
Figure 11 shows the comparison of selfinductance of phase A obtained from experiment and FEA simulation. The experimental result agrees with the simulation one very well, and the maximum error is about 8.2% which is mainly due to the manufacturing imperfection and end effect. It also illustrates that the selfinductance of phase A is approximately constant, so the rotor magnetic circuit is almost symmetric.
The efficiency of the proposed motor is measured at different load currents and outerrotor speeds. As shown in Figure 12, the measured efficiency is higher than 75% over a wide load range at different speeds, whereas the maximum efficiency is about 80% which occurs at 200 rpm. The power losses of motor include copper losses, hysteresis losses of the core, eddy current losses of the PM, mechanical losses, and stray losses. Assuming the armature current is sinusoidal, it has the same frequency and synchronizes with the back EMF. A 2D FEA model is built to analyze the losses of the motor at 400 rpm and fullload. It turns out that the copper losses are 13.32 W. The hysteresis losses of the stator core, rotor core, and modulation blocks in the stationary ring are 2.55 W, 0.72 W, and 1.38 W, respectively. And the eddy current losses of them are 3.66 W, 1.61 W, and 2.46 W, respectively. All the iron losses count up to 12.38 W. The eddy current losses of the PM are 1.6 W. The stray losses and mechanical losses are totally estimated to 10 W, about 2% of the rated power. The simulated losses are less than the experimental ones, mainly because of the neglected end effect, the harmonics of the input current introduced by the power devices in the experiment, the inaccuracy of the iron losses calculation method for not taking the influence of DCbiased magnetic induction into consideration, and the inevitable manufacturing imperfection. It should be noted that the system efficiency of the FMDD motor is relatively lower than a standard direct drive PM machine, because the operation principle of the motor is based on utilizing the harmonic field, which will introduce large amounts of iron losses. Nevertheless, such reduction of efficiency is well outweighed by the uniqueness of selfdeceleration with no gear.
To assess the performance of the proposed motor, the hysteresis current control with = 0 vector control method is implemented. Figure 13 shows the comparison of phase currents between the simulation and experimental results under the rated load of 8 Nm at 400 rpm. It should be noted that phase current is effectively controlled to be sinusoidal and the two results agree well. Nevertheless, the amplitude of the experimental result is slightly higher than the simulation one. It is expected since the simulation result has not taken into account the parasitic losses which require additional current for the desired performance. The output torque waveforms of the experimental and the simulation results are shown in Figure 14. It can be observed that the torque ripple is very low which is due to the fact that phase current is synchronized with the back EMF when the = 0 control is used, and there is a large value of the smallest common multiple between stator slot number and the rotor pole pairs. The slight difference between them is mainly caused by the discrepancy of damping torque in the simulation and the experiment. Moreover, the measured output torque of 8.2 Nm validates the design of modulation block ratio for the stationary ring. The measured torque density is about 9 kNm/m^{3}; it is lower than a standard direct drive PM machine; this is primarily due to the manufacturing imperfection and no optimization of the machine design. The maximization of the torque density will be addressed in the future work by employing concentrated, single tooth windings and fractional numbers of slots/pole/phase, with high slot fill factors and short endwindings.
Finally, the torquecurrent and torquespeed characteristics are measured as shown in Figures 15 and 16, respectively. Figure 15 shows the torque versus current at the speed of 400 rpm; it shows a linear torque and current relationship. Figure 16 verifies that the proposed magneticgeared PM motor can guarantee a constant torque over a wide speed range. Hence, this FMDD motor can operate in a low speed and provide a steady torque output without using a mechanical gearbox.
6. Conclusion
This paper has designed, analyzed, and implemented a fluxmodulated direct drive motor with two airgaps using the sizing equation and FEA. The configuration and operating principle of the motor are described. Field distribution, inductance characteristics, and cogging torque of the motor are investigated using the 2D FEA. The evaluation of system performances is conducted by computer simulation. To verify the performance of the proposed motor, an experimental system is built. The back EMF, efficiency, torquespeed characteristic, and torque output have been measured, and good agreement between the simulation and experimental results has been found. Both the simulation and experimental results confirm that thisfluxmodulated direct drive motor can achieve speed changing without the gearbox and provide a steady output torque with a simple control method.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This work was supported in part by the NSFC under Projects 51177012 and 61374125, the 973 Program of China under Project 2013CB035603, and the Natural Science Foundation of Jiangsu Province, China, under Project BK2011458.
References
 K. T. Chau, C. C. Chan, and C. Liu, “Overview of permanentmagnet brushless drives for electric and hybrid electric vehicles,” IEEE Transactions on Industrial Electronics, vol. 55, no. 6, pp. 2246–2257, 2008. View at: Publisher Site  Google Scholar
 D. G. Dorrell, M.F. Hsieh, M. Popescu, L. Evans, D. A. Staton, and V. Grout, “A review of the design issues and techniques for radialflux brushless surface and internal rareearth permanentmagnet motors,” IEEE Transactions on Industrial Electronics, vol. 58, no. 9, pp. 3741–3757, 2011. View at: Publisher Site  Google Scholar
 F. Y. Cheng and C. H. Chao, “Direct drive torque motor,” U.S. Patent 7633199B2, 2009. View at: Google Scholar
 K. Atallah, S. D. Calverley, and D. Howe, “Design, analysis and realisation of a highperformance magnetic gear,” IEE Proceedings: Electric Power Applications, vol. 151, no. 2, pp. 135–143, 2004. View at: Publisher Site  Google Scholar
 P. O. Rasmussen, T. O. Andersen, F. T. Jørgensen, and O. Nielsen, “Development of a highperformance magnetic gear,” IEEE Transactions on Industry Applications, vol. 41, no. 3, pp. 3741–3756, 2005. View at: Google Scholar
 K. T. Chau, D. Zhang, J. Z. Jiang, C. Liu, and Y. Zhang, “Design of a magneticgeared outerrotor permanentmagnet brushless motor for electric vehicles,” IEEE Transactions on Magnetics, vol. 43, no. 6, pp. 2504–2506, 2007. View at: Publisher Site  Google Scholar
 L. L. Wang, J. X. Shen, P. C. K. Luk, W. Z. Fei, C. F. Wang, and H. Hao, “Development of a magneticgeared permanentmagnet brushless motor,” IEEE Transactions on Magnetics, vol. 45, no. 10, pp. 4578–4581, 2009. View at: Publisher Site  Google Scholar
 W. N. Fu and S. L. Ho, “A quantitative comparative analysis of a novel fluxmodulated permanentmagnet motor for lowspeed drive,” IEEE Transactions on Magnetics, vol. 46, no. 1, pp. 127–134, 2010. View at: Publisher Site  Google Scholar
 S. Taïbi, A. Tounzi, and F. Piriou, “Study of a stator current excited Vernier reluctance machine,” IEEE Transactions on Energy Conversion, vol. 21, no. 4, pp. 823–831, 2006. View at: Publisher Site  Google Scholar
 J. Le Besnerais, V. Lanfranchi, M. Hecquet, R. Romary, and P. Brochet, “Optimal slot opening width for magnetic noise reduction in induction motors,” IEEE Transactions on Energy Conversion, vol. 24, no. 4, pp. 869–874, 2009. View at: Publisher Site  Google Scholar
 N. Bianchi and E. Fornasiero, “Impact of MMF space harmonic on rotor losses in fractionalslot permanentmagnet machines,” IEEE Transactions on Energy Conversion, vol. 24, no. 2, pp. 323–328, 2009. View at: Publisher Site  Google Scholar
 Y. Fan, J. Huang, X. Han, X. Fu, and H. Wei, “Design, analysis and sensorless control of a new selfdecelerating permanentmagnet motor,” in Proceedings of the 38th Annual Conference on IEEE Industrial Electronics (IECON '12), pp. 4104–4109. View at: Google Scholar
 M. Aydin, S. Huang, and T. A. Lipo, “Design, analysis, and control of a hybrid fieldcontrolled axialflux permanentmagnet motor,” IEEE Transactions on Industrial Electronics, vol. 57, no. 1, pp. 78–87, 2010. View at: Publisher Site  Google Scholar
 S. Dwari and L. Parsa, “Design of halbacharraybased permanentmagnet motors with high acceleration,” IEEE Transactions on Industrial Electronics, vol. 58, no. 9, pp. 3768–3775, 2011. View at: Publisher Site  Google Scholar
 Z. Q. Zhu and D. Howe, “Influence of design parameters on cogging torque in permanent magnet machines,” IEEE Transactions on Energy Conversion, vol. 15, no. 4, pp. 407–412, 2000. View at: Publisher Site  Google Scholar
 L. Zhu, S. Z. Jiang, Z. Q. Zhu, and C. C. Chan, “Analytical methods for minimizing cogging torque in permanentmagnet machines,” IEEE Transactions on Magnetics, vol. 45, no. 4, pp. 2023–2031, 2009. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2014 Ying Fan et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.