Vous êtes sur la page 1sur 5

ANSYS 2011

A Novel Axial-flux Electric Machine for In-wheel Gearless Drive in


Plug-in Hybrid Electric Vehicles
W. N. Fu, and S. L. Ho
The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong
A novel low-speed axial-flux-modulated motor is presented. The proposed motor breaks the traditional design rule which stipulates
that the pole-pair numbers of the stator and the rotor must be the same. Special iron segments in the airgap are used to modulate the
magnetic field. It has the merit of having high power density at low speed. It can be fitted into very limited space such as inside the
wheel rim of electric vehicles. Compared with those of radial-flux-modulated motor, its manufacturing and assembling process are
simple and easy. Its performance is compared with those of a radial-flux-modulated motor using two-dimensional and threedimensional finite-element methods.
Index TermsAxial flux, electric motor, electric vehicle, finite element method, flux modulation, low-speed, magnetic field, power
density.

I. INTRODUCTION

price is expected to rise rapidly in the long


term. Emissions from gasoline driven automobiles are
among the main causes of global warming and
environmental pollution. One of the best solutions is to
develop hybrid electric vehicle (HEV). For HEV drives, inwheel electric motors mounted in the rear wheel axles have
many advantages in that the front wheels and rear wheels can
form a series-parallel drive combination without special
mechanical coupling. Since the wheels of vehicles run in low
speed and the dimension of electric machines is inversely
proportional to its running speed, conventional direct drive
electric motor is very bulk if the traction motor needs to
produce a reasonable torque over a wide speed range.
Normally, a mechanical gear is needed to reduce the motor
speed. The use of a mechanical gear reduces the motor size,
but additional space is needed for the gear. The mechanical
gear also reduces the energy transmission efficiency. Recently,
magnetic gears (MG) are proposed to compete with
mechanical gears in terms of torque transmission capability
and efficiency [1]. Compared to their mechanical counterparts,
MGs have a highly competitive torque transmission capability
with very high efficiency. The MG can be directly combined
with a conventional permanent magnet (PM) motor inside one
frame [2-4]. The system torque density can be significantly
improved. However such system has two rotating parts. Its
mechanical structure is complex and it runs noisily.
A novel simple magnetic geared motor that integrates MG
with a conventional outer-rotor PM brushless motor is
presented recently [5-6]. According to the reported system, the
MG is integrated with a conventional outer-rotor PM
brushless motor and there is only one rotary part. The outerrotor is equipped with sintered NdFeB magnets. Fig. 1 shows
such a motor. Its stator has a 3-phase concentrated winding
RUDE OIL

which can produce a rotating magnetic field with 3 pole pairs,


and the outer-rotor is equipped with 22 pole pairs. It has
stationary iron segments which are made of silicon steel
laminations to modulate the airgap field space harmonics, and
the rotor can rotate at low speed. The operating principle of
the setup is similar to that of MG. However, the high-speed
rotary field is created by an armature rather than with magnets.
The overall size of the unit is more compact than a motor and
gear combination.
In this paper the idea of flux-modulation is extended to
axial-flux motors. A novel low-speed axial-flux-modulated
motor (AFMM) for the in-wheel gearless drive of HEV is
proposed. The AFMM has a 3-phase concentrated winding
which can produce a rotary magnetic field with 3 pole pairs,
and the outer-rotor has 22 pole pairs. Iron segments in the
airgap are used to modulate the magnetic field. It can operate
with high power density at low speed and hence can be used
as direct drives in electric vehicles. Its manufacturing and
assembling processes are simple when compared with those of
radial-flux-modulated motor (RFMM). With AFMM, the front
wheels and rear wheels can operate as a series-parallel drive
without special mechanical coupling between them. The
advantages of AFMM are:
(1) Because of the space constraints in the wheel, the disc
shape is well suited for direct coupling of the motor with a
wheel. Because the ratio between the airgap diameter and the
axial length of iron cores is large, the axial-flux design can
further boost the torque density significantly.
(2) The manufacture process of AFMM is much simpler
than that of the radial-flux-modulated motor (RFMM). Both
the iron segments and the stator are made from soft magnetic
compound (SMC) materials in modular structures and can be
assembled easily [7-9].
(3) The coils on the two sides of the stator core are wound
back-to-back toroidally in order to shorten the length of the

ANSYS 2011

end windings which share a common back iron, thereby


saving the copper materials and improving the power density.
(4) Because of the small number of stator slots, the slot
space is used efficiently.
(5) It has good heat dissipation because of the naturally
formed ventilating ducts between the iron segments.
II. BASIC DESIGN RULES OF FLUX-MODULATED MOTORS
In flux-modulated motors (FMM), the numbers of pole
pairs of the stator and the rotor are different. Usually the rotor
has PMs and has large number of pole pairs to allow it to
rotate at low speed efficiently. The stator has armature
windings and it has small number of pole pairs and thus the
number of slots can be small. The difference between FMM
and normal motors is that the former has stationary iron pieces
between the stator and rotor. These iron pieces modulate the
magnetic field produced by the stator windings and the
number of pole pairs of the strongest specific high-order
harmonic is the same as that of the PM rotor. That means the
magnetic field produced by rotors PMs and the strongest
high-order harmonic field produced by the stator winding will
rotate at the synchronous speed of the motor. The constant
average output torque is produced by the reactions between
the magnetic fields of the stator and rotor. The theory of
magnetic field modulation is the same as that of magnetic
gears [1][5]. The gear ratio of the stationary iron pieces is:
kp
+ k2 N iron
(1)
Gr = 1 stator
k1 pstator
where pstator is the number of stator pole pairs and Niron is the
number of stationary iron pieces. In (1) k1 and k2 can be 1, 3,
5, , , respectively. If k1 = 1 and k2 = -1, the stator winding
will produce the highest space harmonic field. The number of
rotor pole pairs protor should be Grpstator. Therefore, the
relationship among the number of stator pole pairs pstator, the
number of rotor pole pairs protor and the number of stationary
iron pieces Niron is:
(2)
N iron pstator = protor

Fig. 1. An overview of an intersection structure of a radial-flux-modulated


PM motor (Motor I).

In this paper an AFMM is presented and its performances


are compared with that of a RFMM. In both motors the
number of pole pairs in the stator pstator is 3 and the number of
pole pairs in the rotor protor is 22. The number of stationary
iron pieces Niron is 25. The gear ratio G = 3 25 = 22 and
r

Gr pstator

22
=
3 = 22 = protor .
3

The coil arrangement of the stator

windings is the same as that of a conventional 3-phase motor


with 3 pole pairs. The phases in the stator slots along the
circumferential direction are: A, A, -C, -C, B, B, -A, -A, C, C,
-B, -B. For AFMM, in the opposite side of the core, the
current in each slot will be the same as that in slots on the
other side but with opposite direction. To compare power
densities, the RFMM and the AFMM are assumed to have the
same outside dimensions.
III. PERFORMANCE ANALYSIS OF RADIAL-FLUX-MODULATED
MOTOR
The original RFMM as shown in Fig. 1 is referred as Motor
I. Its design data are listed in Table I. Taking into account of
the limitation in current density (assumed to be 9 A/mm2 in
this study) in the conductors of the stator slots, the magnitude
of the ampere-conductors INslotNconductor is 28758 A, where I is
the phase current, Nslot is the number of stator slots and
Nconductor is the number of conductors in each slot.
The performances of RFMM are analyzed using 2-D timestepping finite element method (FEM) of transient magnetic
field electric circuit mechanical motion coupled model
[10-11]. The FEM mesh is refined until the solution

ANSYS 2011

differences between the final mesh with 27250 triangle


elements and the previously refined mesh with 15448 triangle
elements are within a small tolerance. When this problem is
also solved in 3-D, the differences in solution accuracy are
virtually unnoticeable. The magnetic torque is computed using
the virtual work method [12-13]. The core loss is computed in
time domain using the method presented in [14].
When the rotor is at full load, the computed torque is 58
Nm and the back emf is 37 V. The core loss is 282.3 W. The
magnitude of this motors cogging torque is 0.03 Nm.
TABLE I DESIGN DATA OF THE RADIAL FM MOTOR (MOTOR I)
Frequency
220 Hz
Axial length of iron core
Total axial length
Outside radius of outer-rotor
Outside radius of PM
Inside radius of PM
Outside radius of stationary ring
Inside radius of stationary ring
Outside radius of stator
Inside radius of stator
Number of outer rotor pole pairs
Number of stationary iron pieces
Number of stator pole pairs
Number of stator slots

40 mm
64 mm
92 mm
88.6 mm
80.8 mm
80.2 mm
67.2 mm
66.6 mm
17.5 mm
22
25
3
36

IV. PERFORMANCE ANALYSIS OF AXIAL-FLUX-MODULATED


MOTOR
The structure of AFMM is shown in Figs. 2 and 3. It is
referred as Motor II. The coils on the two sides of the stator
core are wound back-to-back toroidally. The basic structure of
the PM and iron core is similar to those of normal axial-flux
motors. The difference is that between the PM and the stator
iron core, there are iron segments in the AFMM. Its design
data are listed in Table II.
To compare the performance of the two motors, both
machines are assumed to have the same axial length, the same
outside frame radius, the same phase number, the same stator
and rotor pole number as well as the same total PM thickness.
There are 3 phases in the two motors being studied. The
supply frequency is 220 Hz and the rotor runs at 600 rpm.
The basis for comparison on the power density of different
motors is that the temperature rises at full-load are the same
for the two motors. For simplicity, the temperature rises are
assumed to be proportional to the total losses in the machines.
As the core loss is only a small percentage of the total losses
in these motors, it suffices to assume the temperature rises in
both motors are the same if their copper losses are the same.
The values of INslotNconductor for the AFMM and RFMM are
listed in Table III.
The performances of AFMM are analyzed using 3-D timestepping FEM of transient magnetic field electric circuit
mechanical motion coupled model [15]. The FEM mesh is

refined to ensure reliable solutions. The solution differences


between the mesh with 338798 tetrahedron elements and the
mesh with 513987 elements are almost the same. The plot of
magnetic flux density on the surface of rotor iron core is given
in Fig. 4. It shows that at the rotor side, one can still see the
three pole-pair magnetic field produced by the stator windings.
The plot of magnetic flux density on the cross-section of the
x-z plane (z is the axial direction of the motor) is shown in Fig.
5. The object in the center is the central piece of the stator iron
core. It shows that the arrangement of polarizations of the
PMs at the two sides of the stator is south pole to south pole
and north pole to north pole. The torque curve and back emf
curves versus time at full-load are shown in Figs. 6 and 7,
respectively. It is noted that the torque ripple is not very large
and the three-phase back emf waveforms are symmetrical.
The magnitude of the cogging torque is 1.2 Nm.
TABLE II DESIGN DATA OF THE AXIAL FM MOTOR (MOTOR II)
Frequency
220 Hz
Total axial length
Outside radius
Inside radius
Thickness of PM
Thickness of stationary iron
Airgap between PM and stationary iron
radius of stationary ring
Airgap between stationary iron and stator
Number of outer rotor pole pairs
Number of stationary iron pieces
Number of stator pole pairs
Number of stator slots

64 mm
92 mm
60 mm
3.9 mm
6.5 mm
0.6 mm
0.6 mm
22
25
3
36

TABLE III ELECTRIC LOADINGS OF THE TWO DIFFERENT MOTORS


Motor type
Motor I
Motor II
(Radial flux)
(Axial flux)
36
36
Nslot
5
5
Nconductor
I (A)

159.77

124.4

ANSYS 2011

Fig. 4. The plot of magnetic flux density on the surface of rotor iron core
(Motor II).
Fig. 2. An axial-flux-modulated-motor (Motor II) (assembled view).

The performances of the two motors are summarized in


Table IV. The FEM simulations show that the AFMM
produces about 40% higher torque when compared to that of
RFMM. The coreloss of the AFMM is higher than that of the
RFMM because the stator volume of the AFMM is smaller
than that of the RFMM. However, because the AFMM has
shorter end windings, the copper loss is small. Therefore, the
total loss of the AFMM is smaller that those of the RFMM.

Fig. 5. The plot of magnetic flux density on the cross-section of x-z plane
(Motor II).
Fig. 3. An axial-flux-modulated-motor (Motor II) (disassembled view).

Fig. 6. The torque at full-load operation (Motor II).

ANSYS 2011

[6]

[7]

[8]

[9]

[10]
Fig. 7. The induced emf at full-load operation (Motor II).
TABLE IV COMPARISON OF DIFFERENT MOTORS
Motor type
Motor I
Motor II
(Radial flux)
(Axial flux)
Output torque (Nm)
58.0
81.2
Coreloss (W)
Copper loss (W)
Total coreloss and copper loss

59.0
282.3
341.3

98.5
223.3
321.8

V. CONCLUSION
The RFMM can be modeled using 2-D FEM and the
AFMM can be modeled using 3-D FEM. For the in-wheel
drive of HEV, AFMM has higher power density than that of
RFMM. The manufacturing process of AFMM is much
simpler than that of RFMM.
ACKNOWLEDGMENT
This work was supported in part by The Hong Kong
Polytechnic University under Grant B-Q18X.
REFERENCES
[1]

[2]

[3]

[4]

[5]

M. Aubertin, A. Tounzi and Y. Le Menach, Study of an


electromagnetic gearbox involving two permanent magnet synchronous
machines using 3-D-FEM, IEEE Trans. Magn., vol. 44, no. 11, Part 2,
pp. 4381-4384, Nov. 2008.
K. T. Chau, Dong Zhang, J. Z. Jiang, Chunhua Liu and Yuejin Zhang,
Design of a magnetic-geared outer-rotor permanent-magnet brushless
motor for electric vehicles, IEEE Trans. Magn., vol. 43, no. 6, pp. 25042506, June 2007.
K. T. Chau, C. C. Chan, Liu Chunhua, Overview of permanent-magnet
brushless drives for electric and hybrid electric vehicles,
IEEE Trans. Industrial Electronics, vol. 55, no. 6, pp. 2246-2257, June
2008.
L. Jian, K. T. Chau and J. Z. Jiang, A magnetic-geared outer-rotor
permanent-magnet brushless machine for wind power generation, IEEE
Trans. Industry Applications, vol. 45, no. 3, pp. 954-962, May-June 2009.
L. L. Wang, J. X. Shen, Y. Wang and K. Wang, A novel magneticgeared outer-rotor permanent-magnet brushless motor, 4th IET
Conference on Power Electronics, Machines and Drives, 2-4 April 2008,
pp. 33-36.

[11]

[12]

[13]

[14]

[15]

W. N. Fu and S. L. Ho, A quantitative comparative analysis of a novel


flux-modulated permanent magnet motor for low-speed drive, IEEE
Trans. Magn., vol. 46, no. 1, pp. 127-134, Jan. 2010.
M. A. Khan, P. Pillay, R. Guan, N. R. Batane, and D. J. Morrison,
Performance assessment of a PM wind generator with machined SMC
cores, IEEE International Electric Machines & Drives Conference, vol.
2, 3-5 May 2007, pp. 1049-1053.
F. Marignetti, Colli, V. Delli, Stefano R. Di and A. Cavagnino, Design
issues of a fractional-slot windings axial flux PM machine with soft
magnetic compound stator, 33rd Annual Conference of the IEEE
Industrial Electronics Society, 5-8 Nov. 2007, pp. 187-192.
F. Marignetti, G. Tomassi, P. Cancelliere, Colli, V. Delli, Stefano R. Di
and M. Scarano, Electromagnetic and mechanical design of a
fractional-slot-windings axial-flux PM synchronous machine with soft
magnetic compound stator, 41st IAS Annual Meeting, vol. 1, Oct. 2006,
pp. 62-69.
W. N. Fu, P. Zhou, D. Lin, S. Stanton and Z.J. Cendes, Modeling of
solid conductors in two-dimensional transient finite-element analysis
and its application to electric machines, IEEE Trans. Magn., vol. 40, no.
2, pp. 426-434, March 2004.
W. N. Fu and S.L. Ho, Enhanced nonlinear algorithm for the transient
analysis of magnetic field and electric circuit coupled problems, IEEE
Trans. Magn., vol. 45, no. 2, pp. 701-706, Feb. 2009.
W. N. Fu, P. Zhou, D. Lin, S. Stanton and Z. J. Cendes, Magnetic force
computation in permanent magnets using a local energy coordinate
derivative method, IEEE Tran. Magn., vol. 40, no. 2, pp. 683-686,
March 2004.
W. N. Fu and S. L. Ho, Error estimation for the computation of force
using the virtual work method on finite element models, IEEE Trans.
Magn., vol. 45, no. 3, pp. 1388-1391, March 2009.
D. Lin, P. Zhou, W. N. Fu, Z. Badics and Z. J. Cendes, A dynamic core
loss model for soft ferromagnetic and power ferrite materials in transient
finite element analysis, IEEE Tran. Magn., vol. 40, no. 2, pp. 1318-1321,
March 2004.
P. Zhou, W. N. Fu, D. Lin, S. Stanton and Z. J. Cendes, Numerical
modeling of magnetic devices, IEEE Trans. Magn., vol. 40, no. 4, pp.
1803-1809, July 2004.

Vous aimerez peut-être aussi