See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309082181

Design and Comparison of Interior Permanent Magnet Motor Topologies for

Traction Applications

Article  in  IEEE Transactions on Transportation Electrification · October 2016

DOI: 10.1109/TTE.2016.2614972

CITATIONS READS
74 7,168

8 authors, including:

Yinye Yang Sandra Milena Castano

McMaster University McMaster University
31 PUBLICATIONS   609 CITATIONS    14 PUBLICATIONS   153 CITATIONS   

SEE PROFILE SEE PROFILE

Anand Sathyan
Eaton
16 PUBLICATIONS   503 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Traction electric machines for hybrid electric vehicles View project

Hybrid, plug-in hybrid and electric vehicles View project

All content following this page was uploaded by Sandra Milena Castano on 27 October 2017.

The user has requested enhancement of the downloaded file.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Design and Comparison of Interior Permanent Magnet
Motor Topologies for Traction Applications
Yinye Yang, Member, IEEE, Sandra Castano, Member, IEEE, Rong Yang, Student Member, IEEE, Michael
Kasprzak, Student Member, IEEE, Berker Bilgin, Member, IEEE, Anand Sathyan, Member, IEEE, Hossein
Dadkhah, and Ali Emadi, Fellow, IEEE
 traction machine market. Both asynchronous and synchronous
Abstract— Traction motors play a critical role in electrified
vehicles including electric, hybrid electric, and plug-in hybrid
electric vehicles. With high efficiency and power density, interior
permanent magnet (IPM) synchronous machines have been
employed in many commercialized electrified powertrains. In this
paper, three different IPM rotor design configurations, which
have been used in electrified powertrains from Toyota, Nissan,
and General Motors, are comparatively investigated. Each
topology is redesigned and improved to meet new design
requirements based on the same constraints. The designed
motors are then compared and comprehensively evaluated for
motor performance, torque segregation, demagnetization,
mechanical stress, and radial forces. The results suggest that the
single V-shape configuration achieves the best overall
performance and is thus recommended as the best candidate.
Index Terms—Electric machines, hybrid electric vehicles,
interior permanent magnet synchronous machines, IPM,
Chevrolet Volt, Nissan Leaf, Toyota Prius, V-shape, Delta shape.
I. INTRODUCTION
H ybrid and electric vehicles, ranging from micro-hybrid to
all-electric vehicles, are emerging as more efficient and
environmentally friendly products of the automotive industry
[1]-[2]. With the aim to decrease the use of fossil fuels, reduce
the tailpipe emissions, and improve the vehicle efficiency and
performance, conventional internal combustion engine (ICE)
powertrains are upgraded or replaced with electrified
powertrains [3]-[6]. Electric traction motors are among the key
enabling technologies contributing to this transition and they
are in high demand due to their high power density, wide
speed range, and high efficiency [7]-[9]. It is estimated that
traction machines will see a 17.4% annual growth rate from
2013 to 2023, and the number of electric machines for
electrified vehicles will rise from 44.6 million in 2013 to
147.7 million in 2023 [10].
With the advanced power electronics and modern controls,
AC machines have replaced DC machines and dominated the
This work was supported by the Canada Excellence Research Chairs
Program, the Natural Sciences and Engineering Research Council of Canada
(NSERC), the Automotive Partnership Canada (APC) Initiative, FCA USA
LLC, and FCA Canada Inc.
Y. Yang, S. Castano, R. Yang, M. Kasprzak, B. Bilgin and A. Emadi are
with the McMaster Automotive Resource Center, McMaster University,
Hamilton, ON L8S 4K1 Canada (e-mail: yangy9@mcmaster.ca,
castano@mcmaster.ca, yangr27@mcmaster.ca, kasprm2@mcmaster.ca,
bilginb@mcmaster.ca, emadi@mcmaster.ca).
A. Sathyan and H. Dadkhah are with FCA US LLC, Fiat Chrysler
Automobiles, Auburn Hills, MI 48326 USA (e-mail:
anand.sathyan@fcagroup.com, hossein.dadkhah@fcagroup.com).
2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
1
AC machines are utilized in commercially available electrified
vehicles [11] [12]. Induction machines have the most mature
technology and well established manufacturing techniques as
they have been developed for many decades [9]. Induction
machine technology also offers relatively lower cost and
easier controllability. However, the conductors on the rotor
increases the rotor copper losses and, hence, the cooling
requirement. This leads to a lower operating efficiency in
general, when compared with permanent magnet synchronous
machines (PMSM) [13]. On the other hand, using rare-earth
permanent magnets enhances the magnetic flux density in the
air gap and increases the power- and torque-density of
permanent magnet synchronous machines. Therefore, the
majority of automotive manufacturers choose permanent
magnet synchronous machines as the traction motor in their
vehicles [11] [12].
Permanent magnet synchronous machines can be classified
further as surface-mounted permanent magnet (SPM)
machines or interior permanent magnet (IPM) machines. SPM
machines offer relatively easier manufacturability since the
magnets are adhesively bonded to the rotor surface. However,
this also results in potential flying-off of the magnets during
high speed operation. In contrast, IPM machines provide
stronger rotor robustness as magnets are embedded in the rotor
laminations. In addition, intrinsically higher difference
between q- and d- axes inductances in IPM machine
contributes to higher reluctance torque; thus, providing higher
overall torque in the high-speed field-weakening region,
which is necessary for traction applications [13].
Various IPM topologies have been studied in a number of
papers. Wu et al. compared circumferential, radial, V-type,
and U-type rotor configurations and concluded that the IPM
rotor configuration performs better for non-overlapping
windings [14]. Wang et al. compared one SPM topology and
four single-layer IPM rotor topologies for high-performance
traction machine applications [15]. Chen et al. compared five
in-wheel exterior rotor topologies including one surface
mounted topology, one surface-insert topology, and three
interior topologies (spoke-type, conventional, and V-shape)
[16]. Charih et al. compared different rotor topologies under
similar slot and pole numbers [17]. Sorgdrager and Grobler
studied the influence of magnet size and rotor topology on the
air-gap flux density of a radial flux PMSM [18]. Arumugam et
al. presented comparative design analysis of permanent
magnet rotor topologies for an aircraft starter-generator
machine [19]. Dorrell et al. studied Toyota Prius IPM, spoke-
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 2

type magnet rotor, and induction machine [20]. Liu et al. Table 1. Design parameters of the traction machine
analyzed performance and parameters for IPMs for electric Components Parameters Target
vehicles [21]. However, these papers all focused on the more Lamination thickness (mm) 0.3
conventional topologies and concentrated on the comparisons Stator Maximal stator outer diameter (mm) 250
Maximal active length (mm) 100
among SPM machines, spoke magnet machines, and IPM Axial segments 4
machines. McFarland et al. investigated the efficiency and Rotor
Lamination thickness (mm) 0.3
performance of three permanent magnet machines with Coil pitch 6
different topologies using ferrite magnet materials [22]. Coil connection Series
Winding
Number of layers 1
Recent publication by Yamazaki et al. studied different IPM Thermal class H
rotor topologies [23] [24] focusing on the cross-magnetization Maximal power (kW) 100
Output
effect and the rotor loss reduction. However, a standard design Maximal efficiency (%) 95
process was missing for the presented machines and it may be the powertrain requirement to achieve the maximum machine
difficult to justify the performance comparison for the three operation speed. Torque ripple at the peak torque is required to
topologies. be less than 5%. Lamination stack length, stator outer
This paper focuses on the design and comparison of three diameter, rotor outer diameter, air gap length, and shaft
major IPM rotor configurations that have been utilized in diameter are all equivalent among the motors. In addition, the
traction machine applications: the single V-shape, double V- same stator geometry and winding configurations are
shape, and delta-shape magnet rotor topologies. All rotors employed. The stator has three-phase distributed windings for
have been carefully designed in order to meet the same design 48 slots and 8 poles, which has been used in most of the
targets specified in Section II. Then, Section III outlines the commercial electrified vehicles mentioned above.
design procedure and Section IV presents the analysis results Furthermore, the total amount of the permanent magnet
by using finite element methods (FEM). Comprehensive materials used in each machine is limited to be less than 1.2
investigations in different aspects of the machine design are kg, which targets at 30% of magnet reduction compared to the
performed to compare the machine characteristics and 2012 NISSAN LEAF All Electric Vehicle (AEV) traction
performance. Finally, Section V concludes the findings and motor design since magnet cost takes up a large fraction of the
contributions of this paper. total motor cost [25]. Finally, the output performances of all
three machines are compared at 100 ̊C magnet temperature in
II. DESIGN TARGETS AND CONSTRAINTS performance simulations. Table 1 presents other design
Hybrid, plug-in hybrid, and electric vehicles (HEVs, parameters for reference.
PHEVs, and EVs) are the three major forms of electrified
vehicles. Toyota selected a hybrid electric platform for its III. DESIGN OF IPM TRACTION MACHINES
popular Prius. Nissan Leaf is purely a battery electric vehicle
A. Material Selection
and General Motors designed the Chevrolet Volt as a range-
extended EV and Chevy Spark as an EV [1]. Despite the The selection of materials plays a vital role in meeting the
differences in platforms, almost all the manufacturers selected design specifications, which determines the performance,
IPM machines as the main traction motor since IPM machines dimensions, weight, cost, and lifetime of the machines. In a
provide high efficiency and high power density as well as typical IPM machine, both the rotor and stator cores are
wide speed-range. However, the way permanent magnets are formed out of laminated silicon steel to effectively reduce the
embedded into the rotor laminations varies among different eddy currents and, hence, the iron losses. The properties of the
manufacturers, which considerably influences the machine silicon steel determine its mechanical strength, permeability,
performance. This paper focuses on the design and magnetic saturation, loss density, and thermal expansion. In
comparison of the three major IPM rotor topologies: the single 1.5 Intrinsic at 20°C
Normal at 20°C
V-shape used by Toyota and Lexus in their hybrid vehicles Intrinsic at 180°C
Normal at180°C
such as the Prius, Camry Hybrid, and RX450h; double V-
Residual magnetism, B [T]

shape used by General Motors in Spark EV and the second

1
generation of Chevrolet Volt; and delta-shape used by Nissan
Leaf and Lexus LS 600H.
In this paper, all three machine topologies follow the same
design targets and are subject to the same constraints. Due to 0.5
confidentiality, the targeted rated current, and rated phase il line
Reco
voltage are normalized and defined to be 1.0 per unit (p.u.).
The max overload requirement is set as the peak torque in this
design process, which is normalized as 1 p.u. torque. 0
-600 -500 -400 -300 -200 -100 0
Continuous operation point is set as the half load to optimize Coercive force, H [kA/m]
the output machine efficiency. The base speed is also targeted Fig. 1. Intrinsic and normal curves for a permanent magnet at 20 and 180
at 1.0 p.u. A speed ratio of 4 is targeted in the analysis due to degrees centigrade and an example recoil permeability curve.

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 3

traction applications, non-oriented silicon steel is widely Start process

utilized as the core material due to its low cost, high yield and
Geometry modification process
tensile strength, high saturation flux density, and low core loss
[5]. It is also possible to manufacture very thin laminations Torque-speed curve analysis Magnet material
selection
using non-oriented steel. In this research, 30JNE electrical

Topology Design and Comparison

Demagnetization analysis
silicon steel is selected as the core material for all three rotor Magnet size and
Torque segregation analysis proportion
topologies and stators.

Geometry Optimization Process

Permanent magnets provide the magnetic torque component Mechanical stress analysis Flux barrier
modification
of the IPM machine. They contribute to higher torque density
Preliminary skew angle analysis
when compared with the other types of electric machines such Distance between
magnets
as induction and conventional switched reluctance machines Inductance analysis
(SRMs). Magnets with higher remanence provide more torque V-shape angle
adjustments
while magnets with higher coercivity (absolute) value are less Meet all the
NO
prone to demagnetization. In this design, Neodymium Iron analyses
YES
requirements? Meet torque and NO
Boron (NdFeB) based permanent magnets are selected torque ripple
YES
because of their high remanence and relatively larger coercive requirements?

Design Candidate
force to achieve the best performance. Fig. 1 illustrates typical Preliminary candidate design selection

magnetic characteristics of a high-energy-density NdFeB
permanent magnet. More analysis and discussions about the
selected magnet will be provided in Section IV. In addition,
polyester film and aramid paper are the preferred material for
slot insulation because of their superior thermal and
mechanical properties. For high-speed IPM machines with
high voltage, Mika is often selected due to its high thermal
endurance and good chemical resistance. Meanwhile, copper
is the ideal material for the stator coil in IPM machines due to
its high conductivity, good mechanical strength, and stable
temperature coefficient of resistance.
B. Design process
Fig. 2 illustrates the machine design process. Initial
geometries were generated based on the criteria described in
Section II. 48/8 stator slot and rotor pole configuration is
selected while the machines were designed initially within the
maximal dimension limits. The single V-shape, double V-
shape, and delta-shape designs were created by applying the
same design fundamentals and the same total magnet weight,
whereas each model incorporated different magnet
configurations and positions. Wire gauge and number of turns
were properly selected to achieve 45% of the copper fill
factor. Design changes were made to each of the three motor
topologies in parallel in order to achieve the required
maximum average torque at the rated speed and the minimum
torque ripple before applying rotor skews. Modifications
involved the length and width of the magnets, the positioning
of the magnets on the rotor, the gap between the magnets (if
applicable), the length of the center bridge, and the size and
shape of the flux barriers at each end of the magnets.
Design iterations are evaluated in terms of overall
performance of the machines based on the mechanical and
electromagnetic performance. Torque-speed envelope is
derived under the limit of the maximum current amplitude and
the maximum phase voltage that the DC link can supply.
Output torque, torque ripple, and efficiency are assessed from
the rated speed to the maximum speed, where the trade-off is
made to deliver high output torque and low torque ripple
throughout the speed range.
After repeated refinements and modifications, the candidate
Analysis
Skew angle optimization
Design candidate selection
End process
Fig. 2. Traction machine design flowchart
designs are further evaluated based on a series of simulations
and analysis. Maximum torque-speed characteristic curve is
compared with the design specifications to make sure the
design targets are met. Demagnetization analysis is performed
to check the status of the magnets under various current
loading and temperature loading conditions. Torque
segregation analysis decomposes the output torque into
reluctance torque and magnetic torque by using frozen
permeability technique [26]. Mechanical stress analysis
evaluates the rotor rigidity under high-speed rotor spin to
ensure that the maximum stress on the rotor does not exceed
the yield stress of the lamination steel. For this purpose, a
number of changes and modifications are made to reduce the
stress concentration on the edge of the flux barriers and
bridges. In addition, the air gap flux density, flux linkage, and
inductance are also analyzed and compared for the three
topologies.
In order to decrease the torque ripple percentage of each
topology and meet the minimum torque ripple requirement, a
rotor skew angle optimization analysis was performed. Rotor
skew is a recent trend and technology used for traction
machine applications to reduce the torque ripple. The rotor
lamination stacks are bonded, welded, or bolted together based
on the skew segment height, and then the magnets are inserted
into each lamination segment. The skew angle is achieved by
rotating each segment stack along the rotor axis. Depending
on the rotor length, different number of skew segments can be
selected. In general, high number of skew segments achieve
better torque ripple reduction effects; however, this also adds
more steps in the assembly process and, hence, manufacturing
cost [27]. Therefore, four skew segments have been selected
for this motor design. To simplify the design process,
preliminary skew angles were applied to each segment in the
three topologies in order to evaluate and compare the

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 4

performance. The best performing topology was chosen to

move forward with more detailed skew angle optimization
process to finalize the skew angles and meet all of the
performance requirements. A program has been developed to
perform multi-objective optimization such that the maximized
average output torque and the minimized torque ripple are
both achieved by selecting from the Pareto-front of the
(a) Delta-shape (b) Double V-shape (c) Single V-shape
optimization results [28] [29]. Since the skew angle design
and optimization is not the focus of this paper, more Fig. 3. Full motor views of the delta-shape, double V-shape, and single
V-shape design topologies.
information and details for the optimization of the skew angle
can be found in [27]. layer of the V-shape magnets consists of a different V-shape
C. Design candidates angle, different magnet sizes, and different center bridge
dimensions. The upper V-shape magnets are smaller in overall
Fig. 3 presents the three design candidates: the single V-
size and are positioned with a wider opening angle between
shape, the double V-shape, and the delta shape topologies. In
them. This was found to have the best improvement on
order to simplify the simulation process and reduce the
maximizing the average torque while keeping a minimal
computational complexity, either one-eighth or quarter models
torque ripple. All the corners in the flux barriers were rounded
are used as divided by the solid line in Fig. 3, where periodic
in order to reduce the mechanical stress concentration and to
boundary conditions are applied due to magnetic symmetry.
decrease the torque ripple.
The comparison analyses are performed by finite element
3) Delta-shape
analysis (FEA) using commercial multi-physics simulation
The delta-shape design combines a V-shape and a straight
package JMAG including both electromagnetic simulations
bar magnet configuration with a total magnet weight of 1.139
and mechanical simulations.
kg, which is the same as the weight of the single V-shape
1) Single V-shape
design. Among the three magnets per pole, the thickness of the
The single V-shape magnet rotor design is between a
upper level straight magnet is smaller than the lower V-shape
surface magnet machine and a spoke-type magnet machine
magnets. Compared with the single V-shape design, the delta-
[13]. In the design process of the single V-shape machine, two
shape design achieves higher peak torque output due to the
permanent magnets are kept in a V-shape layout while the
addition of the upper straight permanent magnets. The second
magnet mass has been minimized (1.139 kg) to ensure the
layer of magnets also reduces the d-axis inductance resulting
lowest permanent magnet cost. The bridge width between the
in a larger difference between q-axis and d-axis inductances.
magnets is optimized so that it limits the amount of flux
This leads to producing higher reluctance torque. However,
leakage and maintains the required mechanical strength. An
higher terminal voltage is generated as well especially in the
extended bridge is used to better allocate the stress distribution
high-speed range. All the permanent magnets have rectangular
on the rotor, which will be discussed further in Section IV.
surfaces in order to minimize the manufacturing cost during
Accordingly, large surface area flux barriers are used and
mass production. A series of simulations have been carried out
rounded corners are applied to avoid mechanical stress
to optimize and improve the performance of the machine in
concentration on the inner bridge. Lamination cutouts have
order to meet the design targets. It is concluded that the size
been placed in the rotor back iron to reduce rotor weight. It
and relative position of each magnet, the pole-arc-to-pole-
should be noted that the single V-shape magnet machine could
pitch ratio, and the flux barriers of each permanent magnet are
generate high torque output in a wide speed range; however,
all critical parameters that significantly affect the delta-shape
any variations of the flux barrier shape/dimensions, bridge
candidate performance. A specific geometry with minimum
width between the two magnets, and width/arc of the
permanent magnet usage has been selected that meets both the
permanent magnets can result in variations in the torque
rated torque and torque ripple requirements.
production. For that reason, the candidate machine was
subjected to numerous structural changes in order to balance
the trade-off between the torque ripple and the average torque IV. COMPARATIVE ANALYSIS
output. Further details of these changes are outlined in Section A. Magnetic analysis
IV. Since the three machine topologies are designed with the
2) Double V-shape same stator and winding configurations, the magnetic field
The double V-shape design consists of two layers of V- generated by the stator winding is identical under the same
shape magnets for a total of four magnets per rotor pole. The armature current, i.e. same stator magnetomotive force
total mass of the magnets used in the double V-shape topology (MMF). The variations in terms of magnetic fields are majorly
is 1.15 kg, less than the design requirement. The resulted from the difference in magnet configurations. Fig. 4
manufacturability of this type of rotor is more complex (a) compares the air gap flux density of the three rotor
compared to the other two topologies due to the higher topologies under the no-load condition. It can be observed that
number of magnets each with a thinner width constrained by delta-shape magnet provides a wider magnetic pole arc angle.
the same minimal amount of overall magnet weight. Each In addition, double V- and delta-shape magnets provide

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 5

1.5 Table 2. Magnetic flux density and flux lines comparison

Single-V
Double-V
1 Delta No- load Full load θ=50̊
Flux density (T)
0.5

0
Single
-0.5 V

-1

-1.5
0 10 20 30 40 50 60 70 80 90
Mechanical angle (degree)
(a)
Double
V

Delta

(b)
10
Single V Delta shape topology has slightly higher back EMF compared
8 Double V
Delta to the other two.
6
In addition, Table 2 compares the contour plots of the
4
magnetic flux density and flux lines of the three
Torque (Nm)

2
0
topologies under the no-load and full load conditions. It
-2
can be observed that under the no-load condition, the
-4
magnetic fields have minimal differences between the
-6 three design topologies. This is because, without the
-8 stator magnetic field interference, the rotor magnetic
0 10 20 30 40
Rotational angle (degree) field generated by the magnets is determined by the total
(c) amount of magnet materials, which is the same in all
150
Single V
Double V
three topologies. Under the full load condition, due to
100 Delta the interaction between the stator and rotor magnetic
50 fields, highest magnetic flux density occurs on the teeth
that are positioned between the two rotor poles. In
Back EMF (V)

0 addition, higher cross-magnetization effects caused by

-50
-100
-150
0 20 40
Rotational angle (degree)
(d)
Fig. 4. Air gap flux density under no-load condition at 1.0 speed, (a) flux
density waveform, (b) harmonics, (c) cogging torque, and (d) back-EMF
slightly higher magnetic flux density in the air gap due to the
additional magnet layer. This is also evident in the frequency
domain by comparing the fundamental component in Fig.
4(b). However, Fig. 4(b) also shows higher harmonic contents
that results in higher cogging torque from the double V- and
delta-shape magnets at no load, which can be also seen in Fig.
4(c). In addition, Fig. 4(d) compares the back EMF profiles.
the armature reaction are observed on the single V-shape
topology. Here, the leading side (left side in this case) of
the rotor entering the stator magnetic field experiences
higher magnetic flux density than the trailing side (right
60 80 100
side). In the double V-shape and delta-shape magnet
configurations, the front layer of the magnets located
closer to the air gap serves as an extra layer of flux
barriers. This results in higher d-axis reluctance, so that
rotor magnetic flux is more symmetrically distributed
and the cross-magnetization effect is reduced.
B. Torque-speed performance
Fig. 5 plots the torque-speed characteristics for all three
rotor topologies together and the design target. The machine
with the delta-shape magnet generates the highest torque at the

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 6

1.2 Table 3. Torque ripple comparison (without skew)

Target
Single-V Torque 1.0 p.u. 2.0 p.u. 3.0 p.u. 3.3 p.u. 4.0 p.u.
1 Double-V ripple speed speed speed speed speed
Delta
Single V-
14.63 16.24 27.27 34.59 68.19
0.8 shape (%)
Double V-
9.018 26.57 46.69 67.11 -
Torque [p.u.]

shape (%)
0.6
Delta-shape
10.36 16.80 151.68 - -
(%)
0.4
geometry optimization. Geometry optimization focuses
primarily on the rotor and magnet geometry in order to reduce
0.2
torque ripple.
1. Dimensions of the magnets: The length and width of the
0
0 0.5 1 1.5 2 2.5 3 3.5 4 magnets affect the total magnet weight. The ratio between
Rotational Speed [p.u.] length and width changes the flux distribution, maximum
Fig. 5. Torque-speed characteristics of the three design topologies. average torque, and the torque ripple.
2 2. Position of the magnets: The position of the magnets
Single-V
affects the flux linkages as well as the flux leakage. The closer
1.75 Double-V
Delta the magnets to the outer rotor bore are, the higher are the
1.5 resulting maximum average torque and the torque ripple.
Voltage (p.u.)

3. Angle between the magnets: The angle between the two

1.25
1 ripple as the magnets are moved closer to or further from the
0.75
0.5
0 10 20 30
Current excitation angle (degree)
Fig. 6. Phase induced voltage characteristics of the three design topologies
(at 3 p.u. speed).
constant torque region; however, its torque is compromised in
flux weakening region due to the DC link voltage limit. In
comparison, the single V-shape and double V-shape machines
have relatively good torque-speed performance throughout the
entire speed range. The difference can be explained by the
induced phase voltage. As rotational speed increases, the
induced terminal voltage increases, which limits the phase
winding to draw current from the DC link at high speed. In
order to extend the speed range, field weakening is applied at
high speed such that more d-axis current is used in armature
current which suppresses the rotor magnetic field and hence
reduces the induced voltage. Fig. 6 analyzes the induced phase
voltage in the three topologies at the speed of 3 p.u. It can be
observed that, at high current excitation angles, which are
typically applied at high speed operations, the delta-shape
magnet generates the highest phase terminal voltage while the
single V-shape magnet generates the lowest. Therefore, the
machine with the delta-shape magnet is constrained by the
maximum DC link voltage first, while the single V-shape
machine has the largest speed range among the three
topologies.
C. Torque ripple reduction
Excessive machine torque ripple causes unwanted noise and
vibration in the vehicle powertrain and affects the durability of
the electric machine as well as the powertrain. Reducing the
machine torque ripple can be performed initially through the
magnets changes the maximum average torque and the torque
rotor outer bore. The performance is also impacted as the
angle changes the size of the flux barriers, distances between
the magnets, and the length of the center bridge.
40 50 60 70 80 90 4. Width of the center bridge: Flux leakage in the center
bridge occurs when the flux cannot flow up to the stator poles.
Instead, leakage flux flows in a circular pattern around the
areas of the magnet tips. This leads to a reduction in the
maximum average output torque. Decreasing the width of the
center bridge reduces the flux leakage as the bridge saturates.
However, this can also cause higher torque ripple as more
magnetic flux from the rotor is forced to the air gap.
5. Flux barriers around the magnets: Creating flux barriers
at each end of the magnets on the rotor helps to adjust air gap
flux density harmonics. Properly designed flux barriers enable
both higher output torque with limited flux leakage and lower
torque ripple.
6. Rounded edges on flux barriers: Rounding the corners on
the flux barriers helps to reduce the harmonics as well as the
mechanical stress in those areas of the rotor.
Table 3 compares the torque ripple percentages of each
topology after the geometry optimization at speeds from 1.0
p.u. to 4.0 p.u. At the base speed, the double V-shape provides
the smallest torque ripple, however, its torque ripple increases
significantly as the speed increases. For high speed operation,
the double V-shape machine failed to meet the maximum 4.0
p.u. speed requirement and the delta-shape machine could not
operate above the 3.3 p.u. speed. It should be noted that these
values are obtained before applying skew angle optimization,
which will greatly reduce the final torque ripple values.
D. Efficiency
Since these three topologies share the same stator geometry,
winding configuration, phase current, and DC link voltage, the
efficiency is determined by the output power, core loss, and

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 7

Table 4. Loss and efficiency comparison among the three topologies Table 5. Demagnetization comparison among three topologies

Rotatio Stator Rotor Coil Mech Magnet temperature

Effic Single V Delta Double V
nal core core copper anical (C)
iency
speed loss loss loss loss
(%) Demagnetization Ratio
(p.u.) (W) (W) (W) (W) Contour Plot: %
100.00
1 413.7 58.1 6873.6 35 93.5 75.00
Single 50.00
3 1583.4 263.6 2483.1 315 92.23
V- 25.00
3.3 1753.8 283 2199.6 388.9 90.18
shape 110 0.00
4 2213.8 350.8 1933.2 560 90.12 Demagnetization Ratio
1 371 52.9 6873.6 35 93.51 Contour Plot: %
Double 100.00
3 1319.8 301.8 2483.1 315 91.71
V- 75.00
3.3 1426 345.5 2064.2 388.4 89.64
shape 50.00
4 - - - - - 25.00
1 384.7 46.94 6873.6 35 93.7 140 0.00

Delta 3 951.8 276.3 2483.1 315 65.43 Demagnetization Ratio

shape 3.3 - - - -
4 - - -
magnet loss, while the copper loss and mechanical loss are
equal in the same operating conditions. For each operating
point, the output torque, rotor core loss, stator core loss are
obtained through finite element analysis (FEA). The FEA uses
core material loss lookup tables in W/m3 and material
conductivity to determine the amount of loss that is generated
by the magnetic flux density, which varies with the rotational
speed and load conditions. The copper loss and mechanical
loss are calculated through analytical and empirical equations
using armature current, winding resistance, and rotational
speed. The magnet loss is neglected as the eddy current in the
magnets is very low due to the segmentation of the magnets
with the skew. More detailed loss calculations of interior
permanent magnet machines can be found in [30] [31]. The
efficiency is calculated by dividing the output mechanical
power by the input power. The output power is calculated by
multiplying the output torque with the speed, while the input
power is calculated by adding all the losses to the output
power. Table 4 presents the loss and efficiency comparison
after the rotor geometry optimization of the three topologies
from the base speed to the maximum operating speed. It can
be observed that the three topologies have similar losses at the
rated point, which are relatively small compared with the
output power. Hence, the efficiency of the three topologies at
the rated point are similar as they all meet the peak torque
requirement. The single V-shape topology has high efficiency
throughout the speed range and meets the maximum 4.0 p.u.
speed. However, both the double V-shape and the delta shape
topologies fail to meet the maximum speed requirement, and
the efficiency starts to drop after 3.0 p.u. speed as the output
torque drops.
E. Demagnetization analysis
Demagnetization analysis is performed to compare the three
topologies at temperatures of 110°C, 140°C, and 200°C,
currents varying from 0.2 p.u. to 1 p.u., and current excitation
angles varying from 0 to 75 degrees. Armature current is
applied and removed after two electric cycle. Demagnetization
ratio defined in (1) is compared before and after current
excitation at different temperatures:
Contour Plot: %
- 100.00
- - 75.00
50.00
25.00
200 0.00
 B2 
Demagnetization ratio (%) 100   1   (1)
 B1 
where B1 is the magnet residual flux density before current
loading and B2 is the magnet residual flux density after the
load.
If no demagnetization happens, the demagnetization ratio is
0, which means that the magnet has the same capability before
and after the load was applied. However, if the magnet
encounters strong reverse exterior magnetic field and works
below the knee point of its non-linear characteristic curve, the
magnet is irreversibly demagnetized even if the load is
released. Depending on the load conditions, the magnetic B-H
characteristic curve drops to a lower recoil curve (see Fig. 1).
This results in lower magnetic flux density and, hence,
degraded machine performance. Table 5 presents the
demagnetization analysis results under various loading and
temperature conditions. The purple color represents no
demagnetization and the red color is 100% demagnetization.
Among those three topologies, delta- and double V-shape
magnets were demagnetized at 140 °C while single V-shape
magnets not. Hence, the single V-shape topology has the best
demagnetization performance because of its relatively thicker
magnets compared with delta- and double V-shape topologies.
F. Torque segregation analysis
The total torque in IPM is composed of two components:
magnetic torque and reluctance torque [5]. The reluctance
torque is introduced due to variations between the d- and q-
axis inductances because the embedded magnets create flux
barriers in the rotor lamination. By using frozen permeability
technique [26], the torque components can be segregated at
each operating point. Since the volume of the magnets and the
armature current are the same, the magnetic field variance
generated from different rotor and magnet topologies can be
compared as different rotor pole and magnet configurations
yield different magnetic flux paths. Fig. 7 compares the torque
components of the three machine topologies under different

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 8

0.2
Single-V Speed = 4.0 p.u Speed = 4.67 p.u
Total torque Double-V
Delta
σ=319 MPa σ=435 MPa Rotor island
0.15 Magnetic torque d max=72 µm Upper d max=98µm
Torque (p.u.)

bridges

0.1 (a)
Reluctance
torque
0.05

0 Central bridge Rotor back

0 10 20 30 40 50 60 70 80 90 100
iron
Current excitation angle (degree)
(a) Current excitation at 0.2 p.u.
1.4
Single-V
1.2 Double-V
Delta
Total torque (b)
g

1.0
Torque (p.u.)

0.8 Reluctance Width of the

torque central bridge Central bridge
0.6

0.4 Magnetic torque

Upper bridge between
0.2 magnet pocket and
rotor surface Rotor OD
0
0 10 20 30 40 50 60 70 80 90 100
(c)
Current excitation angle (degree)
Fillet
(b) Current excitation at 1 p.u.
Fig. 7. Torque segregation comparison of the three machine configurations.

armature currents. It can be observed that, in the constant
torque region where the current excitation angle is small, the
magnet torque is dominant and the differences between the
total torque among the three topologies are small. However,
reluctance torque increases and contributes to a larger portion
of the total torque after 15° current excitation angle. Due to
the double-layer and straight bar magnet configuration of the
delta-shape magnet machine, the d-axis inductance is lower
than those in the single- and double V-shape magnet
machines, thus, contributing to the larger difference between
q- and d-axis inductances and, hence, larger torque.
Unfortunately, this torque increase can hardly be utilized as
the induced phase voltage increases as shown in Fig. 6. This
limits the torque at the highest speed. In contrast, the single V-
shape machine has the largest d-axis inductance and, thus, has
the lowest torque output under high armature current.
G. Mechanical stress analysis
Mechanical stress analysis evaluates the rotor stress
distribution and associated displacement at high speed
operation [32]. This involves the analysis of the centrifugal
force that depends on the geometry of the rotating structure
and the rotational speed. Higher speed impacts the centrifugal
force in a quadratic relationship, which significantly increases
the rotor stress distribution. In traction machines, high speed
operation is typically desired as a design target. The only way
to reduce the mechanical stress is to reduce the rotating mass
and modify the rotor structure so that the local stress can be
distributed. This is not a simple task since any geometry
change may result in a significant impact on the output torque
and torque ripple [33].
The first design consideration is the width of the flux
bridges. Machines must be designed in such a way that
Fig. 8. Stress distribution and displacement in the single V- shape
magnets at 4.0 p.u and 4.67 p.u speeds, (a) general view, (b) inner bridge
view and (c) upper bridge view.
sufficient magnetic saturation can be achieved in the bridges
to minimize the magnet leakage flux. However, reducing the
bridge width too much can compromise the mechanical
rigidity. Secondly, the mechanical stress is not uniformly
distributed in the rotor surface. High stress typically
concentrates in three areas: (a) the upper bridges between
magnet pocket and rotor surface; (b) the center bridge in the
rotor back iron; and (c) rotor inner surface at the shaft
diameter. Fig. 8 illustrates the stress distribution and
displacement for the selected V-shape topology. Displacement
caused by the centrifugal force results in deformation and
when the stress exceeds certain material strength limits,
deformation will be permanent. Therefore, the mechanical
design process needs to maintain a balance for increasing the
mechanical strength without compromising the torque
performance. The following procedures were applied to the
rotor geometry to improve the mechanical rigidity:
1. Fillets are added at the corners of bridges. Flux barrier
corners are rounded to reduce stress concentration.
2. A wider inner bridge between the V-shape permanent
magnets improves the mechanical strength of the system;
however, it may also affect the output torque performance
[34].
3. The length of the bridge between magnet flux barriers is
adjusted. Radially extending the length of the central bridge
may improve the stress distribution, rather than increasing the
maximum mechanical stress.
4. Adding cut-outs to the rotor back iron reduces the rotor
mass without significantly changing the output torque.
However, depending on the shape, location, and the number of

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 9

Full Load No load Full Load No load Full Load No load

Table 6. Mechanical stress and displacement comparison 90 90
90
1202.5 MPa 60 120 3 MPa 60 120 2.5 MPa
2 2
Von Mises Displacement 2
1.5
Speed (p.u.) 150 1.5 30 150 30 150
Stress (MPa) (mm) 1 1 1 1

180 0 180 0 180 0

4.0 421 0.0719
Single V-shape
4.67 578 0.0978 330 330
210 210 210 330
Double V- 4.0 568.88 0.0541
240 300 240 300
shape 4.67 774.31 0.0737 270 270
240
270
300
(a) Radial force density distribution in (b) Radial force density distribution in (c) Radial force density distribution in
4.0 502.62 0.0697
Delta shape single V-shape machine under no load double V-shape machine under no load Delta-shape machine under no load and
4.67 684.12 0.0949 and full load and full load full load
Half load No load Half load No load Half load No load
90 90 90
120 60 120 1.5 MPa 60
the cut-outs, local stress may be resulted. 120 1.5 MPa 60 1.5 MPa
1 1
1
5. Holes can be added to the rotor island, appropriately. 150 30 150 30 150
0.5
30
0.5 0.5 0.5

This allows the distribution of the rotor stress and reduces the 180 0 180 0
180 0
stress in the central bridge.
The stress distribution and rotor displacement are simulated 210 330 210 330
210 330

at maximum and overdrive speeds of 4.0 p.u. and 4.67 p.u., 240 300 240
270
300 240 300
270 270
respectively. It is noted that von Mises stress is used for (d) Radial force density distribution in (e) Radial force density distribution in (f) Radial force density distribution in
single V-shape machine under no load and double V-shape machine under no load Delta-shape machine under no load and
comparison and temperature effects were neglected in these half load and half load half load

simulations. In all the cases, stress and displacement increase Fig. 9. Radial force density distribution comparison at 1 p.u. speed full load
with the speed, but the relationship between them is more (a)-(c), and 3.3 p.u. speed half load (d)-(f)
complex as they are influenced by the rotor geometries as
well. The maximum displacement does not have direct
proportional relationship with the maximum stress. The
maximum displacement occurs in the rotor island, whereas the
maximum stress occurs in the center and upper bridges. In
comparison, topologies with double layer magnets (i.e. double
V-shape and delta-shape) have shorter central bridges, which
yield lower displacement in the rotor island. However the
stresses in the central and upper bridges are higher than those
in single V-shape magnet topology. After the optimization (a)
process, the single V-shape machine has a significant
advantage when compared to the double V-shape and delta-
shape machines, reducing 26.0% and 16.2% respectively of
the stress as shown in Table 6.
H. Radial force analysis
Generally, radial forces are taken into account in the
estimation of vibration and acoustic noise analysis in all types
of machines. A typical way to examine the radial forces is to
analyze the components of the magnetic flux density vector B,
which are obtained from the finite element analysis. The radial (b)
Fig. 10. Radial force density distribution in the frequency
electromagnetic force density σr is dependent on the radial
domain (a) 1.0 p.u. full load (b) 3.3 p.u. half load domain.
component of the magnetic flux density vector Br and it is
of the rotor pole arc where the permanent magnets locate.
expressed using Maxwell’s stress tensor as . To Fig. 10 shows the harmonic contents of the radial force at
simplify the analysis, the tangential and axial components of the rated speed with full load and the 3.3 p.u. speed with half
the flux density are ignored as they are typically much lower load. It clearly indicates that the major harmonic contents exist
than the radial component. Fig. 9 compares the spatial in multiples of the 12th order, which corresponds to the
distribution of the radial force density at the rated speed (i.e., number of slots per pole-pair. The single V-shape topology
1.0 p.u.) with the full load, 3.3 p.u. speed with the half load, reveals the smallest harmonic content compared with the
and no load for the three topologies. It can be observed that double V-shape and delta-shape configurations.
the double V- and delta-shape configurations generate higher I. Discussions on the Topology Comparison
magnitudes of the radial force density spikes. In addition, the
number of the spikes increases for the two topologies because The previous sections compared different aspects of the
of the contribution from both magnet layers. For the single V- three machine topologies, i.e., the single V-shape, double V-
shape configuration, it can be observed that each magnet pole shape, and delta-shape magnet configurations, with the same
in every pole pitch generates two peaks at higher speeds. The amount of permanent magnets and the same stator and
location of these peaks aligns around of the entrance and exit winding configurations. Each topology has its advantages and
drawbacks. All three topologies have similar torque ripple

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 10

1.06
not the focus of this paper and the designs are not at the stage
0.91 30
where it can be applied yet. Work will continue on the single
V-shape topology to consider manufacturing and mechanical

6
0.76
5
25
integration into drive train systems. This still needs to include
the addition of the housing, gear connections, and cooling
0.61 6 8 system into the thermal model to ensure that the cooling
12
Torque(N∙ m)

20
10 capabilities of the motor are sufficient in managing the
0.45
8 temperatures. More results will be published in the future.
15
14
0.30 10
16 V. CONCLUSION
14 12

12 18 10

18
0.15 16
18 20
20
22
This paper presents and compares three IPM rotor
1614
24 topologies for traction machine applications: the single V-
5
0
0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 shape magnet, double V-shape magnet, and delta-shape
Rotational Speed(RPM) magnet configurations with the same amount of permanent
Fig. 11. Torque ripple map of the single V-shape design. magnets and the same stator and winding configurations.
1.06 Design targets and design process are presented and a detailed
95 comparative analysis is introduced. Three candidate topologies
0.91
86

are designed and compared comprehensively. It is concluded

92

that the single V-shape magnet configuration has good torque-

88

90
0.76
94

speed performance, the best demagnetization performance, the

94 9
90

0.61 smallest mechanical stress, and tolerable radial force contents.

5

85
Torque(N∙ m)

96
Therefore, it is recommended as the final design candidate.
95
92

0.45
80
94 95
ACKNOWLEDGMENT
96

0.30 97
This research was undertaken in part, thanks to funding
94

95 75
96

0.15 97
96
0.00
0.50 1.00 1.50 2.00 2.50
Rotational Speed(RPM)
Fig. 12. Efficiency map of the single V-shape design.
performance and output efficiency at the rated point. The
delta-shape magnet configuration offers the highest output
torque at the constant torque region because of the proximity
of the horizontal magnet to the air gap. However, the
horizontal magnet also causes excessive induced back
electromotive force (EMF) limiting the high-speed operation
of the machine and it fails to meet the high-speed design
target. The double V-shape machine has good torque
performance and speed range. It also offers the least radial
force content. However, it starts to demagnetize at 140 °C and
has the highest mechanical stress at high speed. The single V-
shape configuration has good torque-speed performance, the
best demagnetization performance, the smallest mechanical
stress, and tolerable radial force contents. Thus, it is selected
as the final design candidate out of the three topologies.
Skew optimization as detailed in [29] has been performed to
further reduce the torque ripple. Fig. 11 shows the torque
ripple map for the single V-shape machine after applying rotor
skew. The minimum torque ripple is now below 5% at the
peak torque region. Fig. 12 plots the efficiency map for the
single V-shape machine in the normalized scale. It can be
observed that the peak efficiency of 97% is achieved between
1.0 p.u. speed and 1.7 p.u. speed at lower to medium torque
region.
In addition, thermal performance is a very important aspect
of electric machine design [31]. However, thermal analysis is
92
90
from the Canada Excellence Research Chairs Program, the
92
Natural Sciences and Engineering Research Council of
3.00 3.50 4.00
70 Canada (NSERC), the Automotive Partnership Canada (APC)
Initiative, FCA USA LLC, and FCA Canada Inc. The authors
also gratefully acknowledge Powersys Solutions for their
support with JMAG software in this research.
REFERENCES
[1] Y. Yang, K. Arshad-Ali, J. Roeleveld, and A. Emadi, "State-of-the-art
electrified powertrains: hybrid, plug-in hybrid, and electric vehicles,"
International Journal of Powertrains, vol. 5, no. 1, pp. 1-29, 2016.
[2] B. Bilgin, P. Magne, P. Malysz, Y. Yang, V. Pantelic, M. Preindl, A.
Korobkine, W. Jiang, M. Lawford, and A. Emadi, "Making the case for
electrified transportation," IEEE transaction on transportation
electrification, vol. 1, no. 1, pp. 4-17, 2015.
[3] Y. Yang and A. Emadi, "Hybrid and plug-in hybrid vehicles," in Wiley
Encyclopedia of Electrical and Electronics Engineering, John Wiley &
Sons, 2013, pp. 1-22.
[4] Y. Yang, N. Schofield, and A. Emadi, "Integrated Electro-Mechanical
Double-Rotor Compound Hybrid Transmissions for Hybrid Electric
Vehicles," IEEE Transactions on Vehicular Technology, vol. 65, no. 6,
pp. 4687-4699, 2016.
[5] A. Emadi, Advanced Electric Drive Vehicles, A. Emadi, Ed., Boca
Raton, FL: CRC Press, ISBN 978-1-4665-9769-3, Oct. 2014.
[6] Y. Yang and A. Emadi, "Integrated Electro-Mechanical Transmission
Systems in Hybrid Electric Vehicles," in IEEE Vehicle Power and
Propulsion Conference, Chicago, IL, USA, 2011.
[7] S. Rogers, "EV everywhere grand challenge, Electric drive status and
challenges," U. S. Department of Energy, Energy Efficiency &
Renewable Energy, 2012.
[8] B. Bilgin and A. Emadi, "Electric motors in electrified transportation: A
step toward achieving a sustainable and highly efficient transportation
system," IEEE Power Electronics Magazine, vol. 1, no. 2, pp. 10-17,
June 2014.
[9] Z.Q. Zhu and D. Howe, "Electrical machines and drives for electric,

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
hybrid, and fuel cell vehicles," Proceedings of the IEEE, vol. 95, no. 4,
pp. 746-765, April 2007.
[10] P. Harrop, "Electric traction motors for EVs see 17.4% CAGR over
next decade," IDTechEx, 05 June 2013. [Online]. [Accessed 10 April
2015].
[11] J. de Santiago, H. Bernhoff, B. Ekergård, S. Eriksson, S. Ferhatovic, R.
Waters, and M. Leijon, "Electrical motor drivelines in commercial all-
electric vehicles: a review," IEEE Transactions on Vehicular
Technology, vol. 61, no. 2, pp. 475-484, Feb. 2012.
[12] A. Bazzi, "Electric machines and energy storage technologies in EVs
and HEVs for over a century," in IEEE International Electric Machines
& Drives Conference (IEMDC), Chicago, IL, 2013.
[13] B. Bilgin and A. Sathyan, "Fundamentals of Electric Machines," in
Advanced Electric Drive Vehicles, Boca Raton, FL, CRC Press, 2014,
pp. 107-186.
[14] S. Wu, L. Tian, and S. Cui, "A comparative study of the interior
permanent magnet electrical machine’s rotor configurations for a single
shaft hybrid electric bus," in IEEE Vehicle Power and Propulsion
Conference, Harbin, China, Sept. 2008.
[15] A. Wang , Y. Jia , and W.L. Soong, "Comparison of five topologies for
an interior permanent-magnet machine for a hybrid electric vehicle,"
IEEE Transactions on Magnetics, vol. 47, no. 10, pp. 3606 - 3609, Oct.
2011.
[16] Q. Chen, G. Liu, L. Sun, Y. Jiang, and J. Yang, "Comparison of five
topologies rotor permanent magnet motors with improved fault-
tolerance," in IEEE International Symposium on Industrial Electronics,
Taipei, May 2013.
[17] F. Charih, F. Dubas, C. Espanet and D. Chamagne, "Performances
comparison of PM machines with different rotor topologies and similar
slot and pole numbers," in International Symposium on Power
Electronics, Electrical Drives, Automation and Motion, Sorrento, 2012.
[18] A.J. Sorgdrager and A.J. Grobler, "Influence of magnet size and rotor
topology on the air-gap flux density of a radial flux PMSM," in IEEE
International Conference on Industrial Technology, Cape Town, Feb.
2013.
[19] P. Arumugam, J. Dusek, A. Aigbomian, G. Vakil, S. Bozhko, T.
Hamiti, C. Gerada, and W. Fernando, "Comparative design analysis of
permanent magnet rotor topologies for an aircraft starter-generator," in
IEEE International Conference on Intelligent Energy and Power
Systems (IEPS), Kyiv, 2014.
[20] D. G. Dorrell, A. M. Knight, L. Evans, and M. Popescu, "Analysis and
design techniques applied to hybrid vehicle drive machines—
assessment of alternative IPM and induction motor topologies," IEEE
Transactions on industrial electronics, vol. 59, no. 10, pp. 3690-3699,
2012.
[21] X. Liu, H. Chen, J. Zhao, and A. Belahcen, "Research on the
performances and parameters of interior PMSM used for electric
vehicles," IEEE Transactions on industrial electronics, vol. pp, no. 99,
2016.
[22] J.D. McFarland, T.M. Jahns, and A.M. EL-Refaie, "Performance and
efficiency comparisons for interior PM and flux-switching PM
machines with ferrite magnets for automotive traction applications," in
IEEE Energy Conversion Congress and Exposition (ECCE), Montreal,
QC, 2015.
[23] K. Yamazaki and M. Kumagai, "Torque analysis of interior permanent-
magnet synchronous motors by considering cross-magnetization:
variation in torque components With permanent-magnet
configurations," IEEE Transactions on Industrial Electronics, vol. 61,
no. 7, pp. 3192 - 3201, Aug. 2014.
[24] K. Yamazaki, Y. Kato, T. Ikemi, and S. Ohki, "Reduction of rotor
losses in multi layer interior permanent magnet synchronous motors by
introducing novel topology of rotor flux barriers," IEEE Transactions
on Industry Applications, vol. 50, no. 5, pp. 3185 - 3193, Jan. 2014.
[25] J. Miller, "Electric Motor R&D," Oak Ridge National Laboratory, U.S.
Department of Energy, 2013.
[26] J.A. Walker, D.G. Dorrell, and C. Cossar, "Flux-linkage calculation in
permanent-magnet motors using the frozen permeabilities method,"
IEEE Transactions on Magnetics, vol. 41, no. 10, pp. 3946-3948, 2005.
2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
11
[27] W. Jiang, Y. Yang, D. Al-Ani, B. Bilgin, A. Sathyan, H. Dadkhah, A.
Emadi, "Torque Ripple Reduction for IPM Synchronous Machines
under Load Excitation," IEEE Transactions on Energy Conversion,
2016.
[28] R. Yang, Y. Yang, B. Bilgin, A. Sathyan, H. Dadkhah, S.C. Veldhuis,
and A. Emadi, "Analysis of Interior Permanent Magnet Machines with
Grain Boundary Diffusion Processed Magnet," in IET International
Conference on Power Electronics, Machines and Drives , Glasgow,
Scotland, UK, 2016.
[29] W. Jiang, B. Bilgin, Y. Yang, A. Sathyan, H. Dadkhah, and A. Emadi,
"Rotor skew pattern design and optimization for cogging torque
reduction," IET Electrical Systems in Transportation, vol. 6, no. 2, pp.
126-135, 2016.
[30] S. Nalakath, M. Preindl, Y. Yang, B. Bilgin, B. Cheng, A. Emadi,
"Modeling and analysis of core losses of an IPM machine for online
estimation purposes," in Annual Conference of the IEEE Industrial
Electronics Society (IECON), Yokohama, Japan, 2015.
[31] Y. Yang, B. Bilgin. M. Kasprzak, S. Nalakath, H. Sadek, M. Preindl, J.
Cotton, N. Schofield, A. Emadi, "Thermal management of electric
machines," IET Electrical Systems in Transportation, 2016, DOI:
10.1049/iet-est.2015.0050.
[32] S. Ramarathnam, A.K. Mohammed, B. Bilgin, A. Sathyan, H. Dadkhah,
and A. Emadi, "A Review of Structural and Thermal Analysis of
Traction Motors," IEEE Transactions on Transportation Electrification,
vol. 1, no. 3, pp. 255-265, 2015.
[33] S. Castano, Y. Yang, B. Bilgin, A. Sathyan, H. Dadkhah, A. Emadi,
"Mechanical stress and radial force analysis in interior permanent
magnet machines," IEEE Transactions on Energy Conversion, 2016.
[34] S. Hahlbeck and D. Gerling, "Design considerations for rotors with
embedded V-shape permanent magnets," in International Conference on
Electrical Machines, 2008.
Yinye Yang (S’11-M’14) received his B.E. degree
in Mechanical Engineering from Tsinghua
University, Beijing, China, M.E. degree in
Mechanical Engineering from University of Illinois
at Urbana-Champaign (UIUC), Illinois, U.S., and
Ph.D. degree in Electrical Engineering from
McMaster University, Ontario, Canada. He then
worked as a principal research engineer for
McMaster University from 2014 to 2015. He joined
Magna Powertrain since 2015 and has been an adjunct professor with
McMaster University since 2016. Dr. Yang is the principal author/co-author
of 5 patent applications, 7 book chapters, and numerous journal and
conference papers. He serves as reviewer and session chair for IEEE
Transactions on Transportation Electrification, IEEE Transactions on
Magnetics, Electric Power Components and Systems, Energies, and IEEE
Transportation Electrification Conference and Expo (ITEC).
Sandra M. Castano (M’15) received the Ph.D.
Degree in in Electronics, Electrical and Automation
Engineering from Rovira i Virgili University,
Tarragona, Spain, in 2010. She has worked on
design, characterization, modeling, simulation and
control techniques of electrical machines for
automotive applications, principally SRM, BLDC
and IPM motors. Since 2013, she is a postdoctoral
researcher at McMaster Institute for Automotive
Research and Technology (MacAUTO) at McMaster
University, Hamilton, Ontario, Canada. Her research interests continue to be
in electrical machine for EV and HEV applications.
Rong Yang (S’15) received the B.S. and M.S.
degree in mechanical engineering from Xi’an
Jiaotong University, Xi’an, China, in 2010 and 2012,
respectively. She is currently pursuing the Ph.D.
degree in the Department of Mechanical Engineering
at McMaster University, Hamilton, Ontario, Canada.
Her research interests include design and
manufacturing issues relevant to brushless
permanent magnet machine design, primarily for
electric vehicle traction application.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2016.2614972, IEEE
Transactions on Transportation Electrification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 12

Michael Kasprzak (S’15) received his Bachelor of
Technology (B.Tech) specializing in Automotive
and Vehicle Technology from McMaster University
in 2014. He is completing his Master of Applied
Science (M.A.Sc.) in Mechanical Engineering at
McMaster University. Currently, he is working as a
research assistant at the McMaster Automotive
Resource Centre (MARC), the Canada Excellence
Research Chair (CERC) in Hybrid Powertrain. His
main research interests include hybrid vehicles,
thermal management of electric machines, and motor design.
Berker Bilgin (S’09-M’12) is the Research Program
Manager in Canada Excellence Research Chair in
Hybrid Powertrain Program in McMaster Institute
for Automotive Research and Technology
(MacAUTO) at McMaster University, Hamilton,
Ontario, Canada. He received his Ph.D. degree in
Electrical Engineering from Illinois Institute of
Technology in Chicago, Illinois, USA. He is
managing many multidisciplinary projects on the
design of electric machines, power electronics,
electric motor drives, and electrified powertrains. Dr. Bilgin was the General
Chair of 2016 IEEE Transportation Electrification Conference and Expo
(ITEC’16). He is currently pursuing his MBA degree in DeGroote School of
Business at McMaster University, Canada.
McMaster University in Hamilton, Ontario, Canada. Before joining McMaster
University, Dr. Emadi was the Harris Perlstein Endowed Chair Professor of
Engineering and Director of the Electric Power and Power Electronics Center
and Grainger Laboratories at Illinois Institute of Technology (IIT) in Chicago,
Illinois, USA, where he established research and teaching facilities as well as
courses in power electronics, motor drives, and vehicular power systems. He
was the Founder, Chairman, and President of Hybrid Electric Vehicle
Technologies, Inc. (HEVT) – a university spin-off company of IIT. Dr. Emadi
is the recipient of numerous awards and recognitions. He was the advisor for
the Formula Hybrid Teams at IIT and McMaster University, which won the
GM Best Engineered Hybrid System Award at the 2010, 2013, and 2015
competitions. He is the principal author/coauthor of over 350 journal and
conference papers as well as several books including Vehicular Electric
Power Systems (2003), Energy Efficient Electric Motors (2004),
Uninterruptible Power Supplies and Active Filters (2004), Modern Electric,
Hybrid Electric, and Fuel Cell Vehicles, Second Edition (2009), and
Integrated Power Electronic Converters and Digital Control (2009). He is
also the editor of the Handbook of Automotive Power Electronics and Motor
Drives (2005) and Advanced Electric Drive Vehicles (2014). Dr. Emadi was
the inaugural general chair of the 2012 IEEE Transportation Electrification
Conference and Expo (ITEC) and has chaired several IEEE and SAE
conferences in the areas of vehicle power and propulsion. He is the founding
Editor-in-Chief of the IEEE Transactions on Transportation Electrification.

Anand Sathyan (IEEE M’ 07) received the B.S. and

M.S. degrees in electrical engineering with highest
distinction from Amrita Institute of Technology and
Coimbatore Institute of Technology, INDIA and the
Ph.D. degree in electrical engineering from Illinois
Institute of Technology (IIT), Chicago, IL under the
guidance of Prof. Ali Emadi. Before joining the
Ph.D. program he was an Assistant Professor at
Amrita Institute of Technology, INDIA. He is
currently a technical specialist in traction motor
design at FCA US LLC, Auburn Hills, MI, USA and an Adjunct Assistant
Professor at McMaster University Canada. Dr. Sathyan is currently serving as
the chair of the IEEE Technical Committee on Vehicle and Transportation
Systems. Dr. Sathyan is the principal author/co-author of several journal and
conference papers. Dr. Sathyan was the member of the Technical Committee
of the 1st IEEE Vehicle Power and Propulsion Conference (VPPCÕ05), and
VPPC 2007 and the IEEE Transportation Electrification Conference (ITEC).
He is the associate guest editor for IEEE transactions on Power Electronics
(Transportation Electrification and Vehicle Systems) and the Associate editor
for IEEE Transactions on Transportation Electrification. He has co- authored
a book chapter in Advanced Electric Drive Vehicles (2014). His area of
expertise is on motor design and motor control.
Hossein Dadkhah received his B.S. degree in
Electrical engineering from the University of Alabama
- Huntsville, AL in 1989, M.S. degree in electrical
engineering and MS degree in engineering
Management from the University of Alabama in 1993
and 1999 respectively. He is currently employed at
FCA US LLC - Auburn hills, Michigan, USA, where
he is responsible for electric machine design,
development and integration for HEV/PHEV and EV
vehicles. He has a wide experience in automotive electronics component and
subsystem design and development and manufacturing process, electrified
powertrain, power electronics and electric motors, audio and telematics
component and system, RF, antenna and tuners, Circuit board design,
EMI/EMC, passive safety components and systems, and electric motor
advance design and modeling.

Ali Emadi (IEEE S’98-M’00-SM’03-F’13) received

the B.S. and M.S. degrees in electrical engineering
with highest distinction from Sharif University of
Technology, Tehran, Iran, in 1995 and 1997,
respectively, and the Ph.D. degree in electrical
engineering from Texas A&M University, College
Station, TX, USA, in 2000. He is the Canada
Excellence Research Chair in Hybrid Powertrain at

2332-7782 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
View publication stats