IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO.
4, DECEMBER 2019
A Review on Multiphase Drives for Automotive
Traction Applications
Ahmed Salem , Student Member, IEEE, and Mehdi Narimani , Senior Member, IEEE
Abstract— This article attempts to cover the most recent
advancements in multiphase drives (MPDs), which are candi-
dates for replacing three-phase drives in electric vehicle (EV)
applications. Multiphase machines have distinctive features that
arouse many research directions. This article reviews the recent
advancements in several aspects such as topology, control, and
performance to evaluate the possibility of exploiting them more
in EV applications in future. The six-phase drives are extensively
covered here because of their inherent structure as a dual
three-phase system, which eases the production process. This
article presents different topologies used in dual three-phase
drives and the modulation techniques used to operate them
as well as the status of using MPDs in traction applications
industrially and the upcoming trends toward promoting this
technology more.
Index Terms— Control, modulation, traction motor drives,
topology.
I. I NTRODUCTION
T HE first MPD was introduced in [1], discussing how
the torque fluctuations in five-phase drives are lower
than the traditional three-phase ones by approximately 67%.
The six-phase case is investigated in [2] as another mile-
stone in the development of industrial MPDs considering
a six-step modulation. The rapid development of industrial
power electronics and modulation techniques of converters
encouraged more research in the area of MPDs. The features
of the MPDs attracted researchers to utilize them in traction
applications such as electric vehicles (EVs) [3]–[6] and electric
ship propulsion [7], [8].
Multiphase drives (MPDs) have splendid features compared
with their three-phase counterparts [9]–[13]. One feature is
splitting the power across a higher number of phases, hence
the power ratings per phase is low. Another feature is the
improved magnetomotive force distribution in the air gap of
the machine which reflects lower torque ripples. Furthermore,
the degrees of freedom are increased based on the number
of independent phase variables of the drive, which provides
vital features such as fault-tolerant operation [14]–[19] and
torque density improvement using harmonic current injec-
tion [20]–[26]. These features of the MPDs are very important
for EV applications and will be reviewed in Section II.
Manuscript received July 19, 2019; revised October 31, 2019; accepted
November 13, 2019. Date of publication November 27, 2019; date of
current version January 7, 2020. This work was supported in part by the
Natural Sciences and Engineering Council of Canada (NSERC), which invests
annually over U.S. $1 billion in people, discovery and innovation, and in part
by the Ontario Research Fund: Research Excellence Program. (Corresponding
author: Ahmed Salem.)
The authors are with the Department of Electrical and Computer
Engineering, McMaster University, Hamilton, ON L8S 4L8, Canada (e-mail:
salema5@mcmaster.ca).
Digital Object Identifier 10.1109/TTE.2019.2956355
2332-7782 © 2019 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.
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
1329
In EV applications, integrating the chargers with the existing
machine is an interesting solution to reduce the cost and
the weight of the vehicle [3]–[6]. The integration is based
on changing the connection of the windings of the machine
and at the same time achieving zero-average torque without
installing additional mechanical brakes. The inherent degrees
of freedom of MPDs can help integrate the on-board charg-
ers with a multiphase machine while achieving zero-average
torque [3]–[6].
MPDs are also studied in marine applications to control
the machine speed instead of using mechanical pole chang-
ing [7]. A five-phase machine with 20 slots and eight poles
combination has an inherent feature of changing pole numbers
from eight to 24 electronically, which is studied in [7].
Another example of using MPDs in electric ship propulsion is
presented and experimentally verified in [8] to test fractional-
slot, six-phase machines to reduce torque ripples and elec-
tromagnetic vibrations to avoid causing damage to motors in
warship applications.
Despite the increasing number of published academic works
related to MPDs, the industrial examples are limited, and
currently, the off-shelf traditional three-phase drives are dom-
inant in the motor drive industry. One of the examples of
manufacturing MPDs is the industrial drives of TM4 for
heavy-duty EVs such as trucks and buses, which are studied
in this article as well. Due to the superior features of MPDs
in terms of power density, performance, and fault-tolerant
capability, it is expected to see more manufacturers produce
MPDs for EV and ship applications. Moreover, the fault-
tolerant capability of MPDs make them very much attractive
for electric aircraft applications.
Another aspect of studying MPDs is modeling, which is
one of the interesting topics to many researchers. The main
theories on modeling the dual three-phase machines are the
vector space decomposition (VSD) [27], [28] and the double
dq model theory [29]. Section III discusses different theories
on decoupling the variables of the drive system as a manda-
tory step in developing proper modulation techniques for the
topologies presented in Section IV.
MPDs comprise power conversion processes, which are
implemented using power electronic converters. One of the
power conversion processes is ac–dc–ac in industrial appli-
cations that are fed from the traditional three-phase mains
using the pulsewidth modulation-based converters. The ac–ac
conversions have been presented in the literature as one solu-
tion using the matrix converters. Another power conversion
is dc–ac conversion, which is applicable in EV applications,
where the source of electrical power is batteries.
 1330 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019

This article covers the recent advancements from the

topological point of view of the converters used in MPDs.
One of the most popular modulation techniques of the two-
level, voltage source inverter (2L-VSI) is the space vector
modulation technique, in which all the possible switching
states of the converter are studied and a volt-second equa-
tion is applied to realise a reference vector by the possible
voltage vectors achieved by the switching states. The five-
phase 2L-VSI is discussed in [30]–[33], the three-level one
in [34]–[38], and the dual-supply topology to supply the
open-end winding configuration of the machine is discussed
in [39]–[43]. As the number of phases (n) increases, the com-
plexity of the modulation technique increases, which makes
the process of developing optimal switching patterns more
tedious. The power electronics converter topologies and their
modulation techniques received some focus, as represented by Fig. 1. Tesla Model S inverter components. 1) Control board. 2) Aluminum
some recent publications in [44]–[49] for seven-phase drives shield. 3) Phase current sensor ferrite ring. 4) Gate driver board. 5) Phase
and in [50]–[54]. These topologies and modulation techniques busbars. 6) Heat sinks. 7) TO-247 package IGBTs. 8) IGBT clips. 9) DC-link
capacitors [55].
will be reviewed in Section IV.
One of the classical control techniques used to obtain the
optimal performance of MPDs is field-oriented control (FOC).
Recently, model-based schemes such as model predictive
control (MPC) was introduced as a simple control scheme. All
these schemes and their features are discussed in the control
section providing a trade-off in the conclusion section between
all of them. In Section V, an overview of different control
techniques employed for MPDs are stated. Section VI deals
with the current status and examples of industrial manufactur-
ing of MPDs for EV applications. Section VII encounters the
trends and challenges in research to improve and incorporate
the MPDs in the EV industry more. Finally, a conclusion is Fig. 2. Phase currents in p.u. for different MPDs.
provided in the last section.
Fig. 2 shows the reduction in the phase current of several
II. M ULTIPHASE D RIVE F EATURES multiphase machines, namely five-, six-, seven-, nine-, eleven-,
and twelve-phase machines. All the phase currents are in
MPDs have unique features compared to the traditional
per unit (p.u.) values, with the base value considered as the
three-phase drives. This section covers the recent research
phase current of the three-phase case. The relation between
contributions that highlight their features.
the number of phases and the per-phase current is inversely
proportional. The five-, six-, and seven-phase cases show a
A. Lower Power/Current Rating per Phase significant reduction in per-phase currents compared to the
three-phase case. The reduction in per-phase current gets
The existing electric motor drive units in the commercial
smaller as the number of phases gets higher, such as on
EVs are designed such that modules or discrete IGBTs are
going from the eleven-phase to the twelve-phase case. The
connected in parallel, as shown in Fig. 1, to deliver the
phase currents of six-phase machines are obviously half of the
load current. Fig. 1 shows the assembly of the Tesla Model currents of three-phase machines, which can be interpreted as
S inverter components. This connection results in different
a chance to choose the same modules used for the existing
problems like the differences in turning on/off the timing due
three-phase drives, and instead of paralleling them to control
to manufacturing differences between the paralleled transis- the same phase, they can be controlled separately to control
tors and different characteristics of the gate driver modules.
different phases.
Another major problem is the different load sharing between
the paralleled transistors. Finally, the heat dissipation of all the
transistors should be managed, so that no transistor exceeds B. Fault-Tolerant Operation
the allowable thermal limit. In MPDs, the power is divided One of the inherent features of MPDs is the fault-tolerant
across a higher number of phases, and each phase is supplied operation. The increased number of phases provides more
from a converter leg using a single component instead of degrees of freedom under faulty conditions. In faulty cases,
using parallel components. Using MPDs solves the problem of MPDs are better than their traditional three-phase counterparts.
paralleled devices at the expense of adding more complexity The performance of the drive is enhanced based on the
to the assembly of the drive as the number of wires increases. optimization criteria discussed in the literature.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS
Fig. 3. (a) Single-neutral connection. (b) Two-neutral connections.
The faulty conditions are the machine winding faults, which
result in the loss of one or more of the phases considered
as open-circuit faults. The fault tolerance capability of the
six-phase drives has been attracting a lot of researchers
recently; however, it is limited to component-level only. The
detection of the fault occurrence is discussed in [56]. Six-phase
machines have different winding configurations, depending on
the space angle between the two three-phase windings. The
machines with the 60◦ angle are named symmetrical six-phase
machines, where the asymmetrical ones are the machines with
a space angle difference of 30◦ . The asymmetrical six-phase
machine, also referred to as the dual three-phase machine, has
advantages in terms of the space harmonics of the fifth and sev-
enth harmonics in the air gap flux of the machine compared to
their symmetric counterparts [57]. The issue of operation under
fault without any hardware reconfiguration for different MPDs
has been investigated for the first time in [19] using current-
source inverters. The criterion for optimization was only
creating a rotating magnetic field with the remaining phases.
In [17], the problem has been investigated for a single,
open-circuit fault using a voltage-source inverter introducing
two different optimization criteria, namely minimum stator
winding losses and a maximum torque. A unified analysis
using the same strategies, but with more than single-phase
winding loss, is discussed in [15]. An improved approach
is introduced in [58] that suggests widening the criteria of
optimization to achieve minimum losses in the full torque
operation range in the case of a single, open-circuit fault. The
study has been extended to be applied to double open-circuit
faults with different possible combinations of the faulty
phases. Baneira et al. [59] investigated the converter losses
in faulty cases after applying different optimization criteria
reported in the literature.
The asymmetrical six-phase machines have two possible
connections of stator phase windings depending on the con-
nection of the neutral point. Fig. 3 shows the possible config-
urations which are two isolated neutral points and a single
isolated neutral point. The single-neutral point connection
proves worthy in the fault-tolerant operation, whereas the two
isolated neutral point connection has better dc-link utilization
and simpler current controllers, because no zero-sequence
currents can physically flow in this type of connection [15].
The following discussion is about the faulty cases for both
asymmetrical and symmetrical six-phase machines.
Fault-tolerant capability is one of the key features of MPDs
to exploit this technology in traction industries. The different
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
1331
faulty cases that the asymmetrical six-phase machine with
a single-neutral connection can operate with optimal current
control are presented in Fig. 4. The faulty phases which are
open-circuited after fault occurrence are colored in red in
Fig. 4. Each case is assigned a notation nX, comprising a
number n representing the number of faulty phases, where
n{1, 2, 3}, and a letter X represents different combinations
of possible faults, where X{A, B, C, D}. For example, two
open-circuit faults could be either from the same set as in case
(2A) or one from each set as in case (2B). Fig. 4 shows all the
possible nonrepeated combinations of the one, two, or three
faulty phases. The other possible combinations are not shown
here because they have the same characteristics. Fig. 5 shows
the maximum postfault torque range in p.u. values under faulty
conditions for asymmetrical machines in (a) and symmetrical
ones in (b). These results are given based on the optimization
discussed earlier in the section.
As can be seen from Fig. 5, the connection of the neutral
points of the machine winding has a major effect on the
postfault torque range. It can be seen that the single-neutral
connection results in higher torque ranges compared to the
two-neutral connection. The two-neutral connection cannot
support optimal current control in some of the cases, such
as three-phase faults for asymmetrical machines and the cases
(2C, 3B, and 3C) for symmetrical ones.
It can also be noticed from the figures that the symmetrical
machines are superior compared to the asymmetrical ones
regarding the torque range under faulty conditions. However,
the authors recommend the asymmetrical six-phase machines
for EV applications for their merits in healthy case condi-
tions, and the difference between torque ranges under faulty
conditions is not high enough to choose symmetrical six-
phase machines instead. However, a hybrid connection of the
neutrals is introduced in [60] to achieve better torque-speed
exploitation by combining the two torque-speed curves with
the single and two isolated neutral connections.
C. Torque Density Improvement
Torque density improvement is another interesting feature
of MPDs that could lead to providing a high-power density
motor drive for EV applications. The variables of multiphase
machines can be modelled by decomposed subplanes, as it
will be discussed in detail in the next section. One of the
subplanes is considered as the torque producing plane and
the others as losses. In [20], it is proven that with a specific
winding layout, the zero-sequence subplane can contribute to
the torque production of the machine.
The concept is that a rotating magnetic field can be produced
by injecting two 90◦ out-of-phase sets of zero-sequence cur-
rents into the two three-phase sets and with the special winding
layout.
The harmonic injection method could increase the torque
density of multiphase motor drives; however, two problems
are associated with the third harmonic injection in six-phase
drives. One problem is providing a neutral path for the
currents to flow through. This problem is addressed in [20]
and the solution is to connect the machine winding neutral to
 1332 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019

Fig. 4. Different faulty cases of asymmetrical six-phase machines.

Fig. 5. Maximum postfault torque range in p.u. values under different faulty cases. (a) Asymmetrical and (b) symmetrical six-phase machines [15].

the midpoint of the dc-link. Bendixen et al. [24] also addressed
the problem of balancing the voltage of the midpoint to prevent
it from drifting during loading using an additional proportional
controller. The other problem is the optimization of the flux
pattern in the machine air gap. During loading of the machine,
the angle of the third harmonic currents varies which can lead
to uneven hills of the resultant total air gap MMF. Several
publications targeted this issue to analyze the problem and
prevent it [61], [62]. The solutions proposed are based on
artificial intelligence techniques like genetic algorithms in [61]
and neural networks in [62].
Wang et al. [21] also studied the optimal third harmonic
amplitude to be injected into the current in a dual three-phase
machine. It is dependent on the ratio of the third-harmonic
back-EMF to the fundamental back-EMF. The optimal third
harmonic into the current in the dual three-phase machine
without third-harmonic back-EMF is one-sixth of the funda-
mental harmonic. For the dual three-phase PM machines with
the third-harmonic back-EMF and two sets of windings shifted
by 30 electrical degrees, the optimal third harmonic into the
current is dependent on the ratio of the third-harmonic back-
EMF to the fundamental back-EMF.
Another proposal of current injection is investigated in [22]
that focuses on injecting harmonic currents of fifth and seventh
orders to produce a flatter top current waveform to increase
the torque per fundamental current ampere. The only problem
is that these harmonic currents are considered as circulating
currents which do not contribute to the torque production. The
optimization of the injected currents and efficiency analysis
have been done in [23]. These techniques can help increase
the torque per fundamental current ampere and provide quite
a good reduction in the overall size and weight of the motor
drives and lead the way to solve one of the main challenges
in the research field of electrified transportation.
Table I shows a summary of the possible methods
to increase the torque density of asymmetrical six-phase
machines via harmonic injection. The constraints on the tooth
and core flux density limit the increase of the produced torque
in induction motors as in [20], which considers machines
with a concentrated winding. The PMSMs are investigated
in [21]–[23] with a possibility to increase the torque den-
sity beyond the values in Table I. The interaction between
the injected harmonic currents and the same components of
back-EMF results in additional torque production. This feature
can be enabled in PMSM by changing the stator winding lay-
out that results in the desired harmonic components of back-
EMF. However, the additional torque improvement comes at
the cost of inevitable increase in torque ripples.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS 1333

TABLE I
C OMPARISON B ETWEEN D IFFERENT H ARMONIC I NJECTION M ETHODS TO I NCREASE THE T ORQUE D ENSITY OF A SYMMETRICAL S IX -P HASE M ACHINES

III. M ODELING OF M ULTIPHASE M ACHINES

To provide a high-performance modulation and control tech-
nique for MPDs, a model of a multiphase machine is required.
In this section, different modeling techniques, especially for
six-phase machines are reviewed. Six-phase machines have
different winding configurations depending on the space angle
between the two three-phase windings. The machines with
the 60◦ angle are named symmetrical six-phase machines,
where the asymmetrical ones are the machines with 30◦ space
angle difference. The asymmetrical six-phase machine, also
referred in the literature as the dual three-phase machine, has
advantages in terms of the space harmonics of the fifth and
seventh harmonics in the airgap flux of the machine compared
Fig. 6. Required dc-link capacitance for different MPDs [63].
to their symmetric counterparts [57].

D. Lower dc-Link Current Ripples A. Double-DQ Model

The dc-link bank of capacitors is one of the most important Based on the double-DQ model, the two three-phase wind-
components in the powertrain of the EV and the power ings in a six-phase machine are treated separately and two
electronic converter. The size of the capacitor bank is about three-phase different (Clarke) transformations are applied to
two-thirds of the overall converter size. The limited lifetime the phase variables for each three-phase winding [29]. The
and degradation with the ageing of the capacitors are also α-axis of the stationary reference frame is aligned with phase-
critical factors that decrease the reliability of the overall sys- a1 of the stator winding. This transforms the six-phase vari-
tem. The study provided in [63] focuses on different types of ables into two sets of stationary reference frame variables,
capacitors that can be deployed in EV applications. The study denoted α1 − β1 and α2 − β2 components, for windings sets
also includes the required dc-link capacitance for different abc1 and abc2 , respectively:
MPDs using Coulomb counting technique to estimate the  t
[ fαj f β j ]t = [Tj ] · f aj f bj f cj (1)
count of charges based on capacitor currents, which can be
calculated from the switching function of the inverter. where f is a general symbol for all the machine variables
Fig. 6 shows a comparison between the required capacitance (voltage, current, or flux) and k is a subscript of value 1 or
in p.u. for different MPDs, considering the base value of the 2 to differentiate between the variables of the two three-phase
dc-link capacitance as the one required for the three-phase windings.
case. This figure shows that the required capacitance decreases The spatial 30◦ displacement between the two windings is
with increased number of phases of the system up to the accounted for in the decoupling transformation. For an asym-
twelve-phase one [52]. metrical six-phase machine, the power invariant transformation
There is a high decrease in the dc-link capacitance when for winding sets abc1 and abc2 , respectively, is given by
compared between three-phase and five-phase cases. The curve ⎡ ⎤
 1 −
1
−
1
is found to be decreasing, considering the cases of six-, seven-, 2⎢
[T1 ] = ⎣ √2 √2 ⎥
and nine-phase drives; however, the capacitance required for
3 3 3⎦
the cases of nine-, eleven-, and twelve-phase machines appears 0 −
to be the same. As can be seen from Fig. 6, the dc-link ⎡√ 2 √ 2 ⎤
 3 3
capacitor values almost reduced to half when a six-phase drive 2⎢ − 0⎥
[T2 ] = ⎣ .
1⎦
is considered, which could help the converter become more 2 2 (2)
3 1
power-dense than a conventional three-phase system. −1
2 2

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 1334 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019

Fig. 7. Topologies of power electronic converters for MPDs.

B. VSD By applying this transformation matrix the variables are trans-

formed into (αβx y0+0− ) reference frame [27]. A summary
Recently, VSD is widely used to model multiphase
comparison between double-DQ and VSD methods is shown
ac machines [64]–[66]. The main purpose is to decom-
in Table II.
pose the machine voltages and currents from the original
six-dimensional space into new independent subspaces. This
transformation provides a full representation of the machine IV. S IX -P HASE P OWER E LECTRONIC C ONVERTERS
dynamics and helps to design a controller with the same AND M ODULATION T ECHNIQUES
structure in healthy and faulty cases. Electrical drives are supplied and controlled recently using
Based on the decomposition theory, all the machine vari- power electronic converters which enable the drive to operate
ables in the original six-dimensional space are transformed in all the possible regions and reach the desired speed and
into three time-orthogonal subspaces. The harmonics of torque. Fig. 7 demonstrates different topologies used for
the original vectors are mapped into three groups, named MPDs that can be exploited in industrial motor applications.
αβ, x y, 0+0− subspaces. The first group represents the fun- All the topologies mentioned in this context are the DC–AC
damental harmonic components and the harmonics of order converters, since the energy storage is mostly battery packs in
k = 12h ± 1 as (h = 1, 2, 3, . . .). The αβ currents are traction applications. The topologies are classified into three
the torque-producing components, since they represent the categories based on the structure and number of levels of
fundamental harmonic. their respective output voltage. The (2L-VSI) topologies are
The other four vectors should be based on their orthogonal- discussed in the first subsection, including the modulation
ity in the aforementioned αβ subspace [27]. It is found that techniques. Since the conventional (2L-VSI) topology is the
the harmonics of k = 6h ± 1 as (h = 1, 3, 5, . . .) spin in the one used in industry, this article focuses more on the modula-
same surface and their subspace is denoted by the x y subplane. tion techniques along with the possible switching patterns. The
The x y current components do not contribute to the torque possibility to reduce the number of switches is also covered
production process, hence they are considered as extra losses in the same subsection featuring the nine-switch converter
in healthy case conditions and as extra degrees of freedom (NSC) and the five-leg, six-phase inverter. Another subsection
in faulty cases. The last group is the triplen harmonics group is dedicated for dual supply inverters used for the open-end
defined by k = 3h, where (h = 1, 2, 3, . . .), and mapped winding configuration, which produces more than two-level
to the 0+ 0− subspace. The current components of the 0+ 0− output voltages based on the same basic (2L-VSI) topology
subspace do not exist, unless there is a physical path for neutral supplying each phase of the machine from both ends. The
currents. The transformation matrix is: final subsection discusses the efforts in the multilevel inverter
topologies, which are promising to be used in the traction
⎡ √ √ ⎤
1 1 3 3 application, since it is expected that future cars will have
⎢1 − − − 0⎥ higher dc-link voltage levels (800 V) [67].
⎢ √2 √2 2 2 ⎥
⎢ ⎥
⎢ 3 3 1 1 ⎥
⎢0 − −1⎥
⎢ 2 2 2
√ √2 ⎥ A. Two-Level Conventional Six-Phase Inverters
⎢ ⎥
1 ⎢ 1 1 3 3 ⎥
[T ] = √ ⎢ 1 − − − 0⎥⎥ (3)
The two-level, six-phase is the most investigated topology
3⎢
⎢ √2 √2 2 2 ⎥ in all the literature of dual three-phase drives due to its
⎢ 3 3 1 1 ⎥
⎢0 − −1⎥
simple structure and availability of industrial packs provided
⎢ ⎥
⎢ 2 2 2 2 ⎥ by manufacturers, as two three-phase bridges and simple
⎢1 1 1 0 0 0⎥ controllers are needed to drive such converters compared to
⎣ ⎦
other complex topologies like matrix, multilevel, and open-end
0 0 0 1 1 1
converters. The structure of the topology is shown in Fig. 8.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS 1335

TABLE II component of each three-phase set to their respective reference

C OMPARISON B ETWEEN THE D ECOUPLING T HEORIES voltages. The method is investigated in several publications
OF A S IX -P HASE S YSTEM
such as the development of the technique in [70], which
shows that it is like the vector classification technique, except
being only simpler to implement as it is carrier based. The
improvement of the harmonic profile is analyzed in [71] and a
general comparison between all the techniques is discussed
in [72] and [73]. A method of designing a proportional-
resonant integrator to control the drive and using the injection
technique has been introduced in [74].
3) Space-Vector PWM (SVPWM) Based on VSD Theory:
Several SVPWM techniques have been presented in the litera-
ture to minimize the circulating currents, which are considered
only as a contribution to the ohmic losses of the drive. The first
SVPWM based on VSD theory was introduced in [27], which
presented the decoupling of the machine variables discussed
in the previous section. Using such transformation enables
the control of the full six-phase converter as a single unit.
Considering all the possible switching states of a six-phase
inverter, there are 64 different switching states, which are
referred to in this article as (V0 : V63 ). The switching states
are sorted in the same way as reported in the literature to
avoid any ambiguity. The states are numbered from 0 to 63 in
a binary manner by considering each upper switch as a digit
in a binary number of the order (Sc2, Sb2, Sa2, Sc1, Sb1, Sa1 )
which makes the first switching state as (000000) and the 64th
as (111111).
Applying these switching states to the inverter, the output
phase voltage can be calculated based on the switching states
from (4) for the two isolated neutral connection and from (5)
Fig. 8. Two-level inverter-fed six-phase machine. for the single isolated one
⎡ ⎤ ⎡ ⎤⎡ ⎤
1) Carrier-Based PWM (CPWM): Like the traditional two- Va1 2 −1 −1 0 0 0 Sa1
⎢Vb1 ⎥ ⎢−1 2 −1 0 0 0⎥ ⎢ ⎥
level, three-phase inverter, the most popular carrier-based ⎢ ⎥ ⎢ ⎥ ⎢ Sb1 ⎥
⎢ Vc1 ⎥ Vdc ⎢−1 −1 2 0 0 ⎥ ⎢
0 ⎥ ⎢ Sc1 ⎥
modulation technique of PWM (CPWM) can be exploited to ⎢ ⎥= ⎢ ⎥
⎢Va2 ⎥ 3 ⎢0⎢ 0 0 2 −1 −1⎥ ⎢ ⎥
operate the six-inverter. The CPWM technique is implemented ⎢ ⎥ ⎥ ⎢ Sa2 ⎥
using the reference signals, which are generated from the ⎣Vb2 ⎦ ⎣0 0 0 −1 2 −1⎦ ⎣ Sb2 ⎦
control scheme and represents two sets of three-phase outputs Vc2 0 0 0 −1 −1 2 Sc2
that shifted by 30◦ . The references are compared to a carrier ⎡ ⎤ ⎡ ⎤⎡ ⎤
(4)
signal, which is obvious from the name of the technique and Va1 5 −1 −1 −1 −1 −1 Sa1
the output signals are used to fire the switches of the converter. ⎢ Vb1 ⎥ ⎢−1 5 −1 −1 −1 −1⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥ ⎢ Sb1 ⎥
The operation of the converter is investigated in [68] and of ⎢ Vc1 ⎥ Vdc ⎢−1 −1 5 −1 −1 −1⎥ ⎢ Sc1 ⎥
⎥ ⎢
⎢ ⎥= ⎢ ⎥.
the high-frequency machines in [69]. CPWM is the simplest ⎢Va2 ⎥ ⎢
6 ⎢−1 −1 −1 5 −1 −1⎥ ⎢ ⎥
⎢ ⎥ ⎥ ⎢ Sa2 ⎥
modulation technique to operate the multiphase converter. This ⎣ Vb2 ⎦ ⎣−1 −1 −1 −1 5 −1 ⎦ ⎣ Sb2 ⎦
simplicity in addition to reliability is the big advantage that Vc2 −1 −1 −1 −1 −1 5 Sc2
makes this technique suitable for EV applications. One of (5)
the disadvantages of CPWM is the lack of control degrees to
achieve a better performance of the converter, such as lowering The transformation matrix (3) is applied to the resultant
the total harmonic distortion (THD) of the output waveforms. output voltages to transfer them to the decoupled subplanes
Another disadvantage is the lower dc-link utilization compared (Vα − Vβ ), (Vx − Vy ), and (V0+ − V0− ). Fig. 9 shows the
to space vector techniques, which can be compensated with projection of all the possible switching states in the three
the zero-sequence injection that will be discussed in the next orthogonal subplanes. Regarding the 0+ , 0− components, all
section. the vectors are mapped to the origin point in the case of
2) Double Zero-Sequence Injection (DZSI): The concept the two isolated neutrals; however, in the single isolated
is based on the infamous zero-sequence harmonic injection neutral point connection, the vectors are mapped, as presented
technique exploited with three-phase inverters to lower the in Fig. 9(c). The vectors in the αβ subplane are classified into
THD of the output and extend the linear control region of the four categories based on their magnitude. Table III shows the
inverter. It can be implemented by adding the zero-sequence four categories with their respective colors plotted in Fig. 9.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 1336 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019

Fig. 9. All possible switching states projected onto three subplanes. (a) αβ subplane. (b) xy subplane. (c) 0+ 0− subplane.

TABLE III vectors of magnitude V1 and one vector of magnitude V3 . The

V OLTAGE V ECTOR C ATEGORIES
The SVPWM is exploited by realizing the rotating reference
determined by the controller. The components of the reference
are the input to (6), which calculates the dwell times of the
possible vectors to be applied:
⎡ ⎤ ⎡ 1 ⎤−1 ⎡ ∗ ⎤ the zero-vector have the notation B2. The 24-sector patterns
T1 Vα Vα2 Vα3 Vα4 Vα5 Vα · Ts
⎢ T2 ⎥ ⎢ Vβ1 Vβ2 Vβ3 Vβ4 Vβ5 ⎥ ⎢ Vβ∗ · Ts ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ period of sampling time and easier implementation, since each
⎢ T3 ⎥ = ⎢ V 1 V 2 V 3 V 4 V 5 ⎥ ⎢ V ∗ · Ts ⎥ . (6)
⎢ ⎥ ⎢ x x x x x ⎥ ⎢ x ⎥ switch is turned on or off one time per period compared to
⎣ T4 ⎦ ⎣ V 1 V 2 V 3 V 4 V 5 ⎦ ⎣ Vy∗ · Ts ⎦
y y y y y the 12-sector patterns which have more than one transition per
T0 1 1 1 1 1 Ts
Here, in Vγk , the superscript k{1, 2, 3, 4, 5} is the sequence
of the applied voltage vector and the subscript γ {α, β, x, y}
denotes the axis to which the voltage vector is synthesized.
Since there are four degrees of freedom in the two separated
neutral connection, four system equations are adopted to
calculate the dwell times. For the single-neutral connection,
it is proven in the literature that the zero-sequence voltage
components are nullified without any further calculations
based on the nature of the voltage vectors mapped to this
subplane, as shown in Fig. 9(c). However, choosing the vectors
that cancel the components of the x y subplane is more
challenging. References stated several switching patterns based
on the number of sectors of the modulation technique. The two
distinctive strategies are 12-sector SVPWM and the 24-sector
SVPWM [63]–[65]. In each sector, four different vectors of
magnitude V1 are chosen to synthesize the reference output
by the volt second concept, such as in (6). The chosen vectors
have a magnitude of V1 in the αβ subplane. On the other
hand, the 24-sector SVPWM is implemented by choosing three
method of inserting the zero-vector determines the continu-
ous and discontinuous modulation scheme, as in [75]–[78].
As stated in the literature, the 24-sector SVPWM method is
easier to implement using DSP microcontrollers and results in
less THD in the output waveforms of the converter.
Table IV compares the reported switching patterns based on
the 12- and 24-sector switching patterns for both continuous
and discontinuous modulations. The name of each switching
pattern starts with the letter C for continuous modulation and
D for the discontinuous one. The patterns that start and end
with the application of a zero-vector have the letter A at the
end of the notation, while the ones the only start with the
zero-vector have the notation B1 and the ones that end with
have the advantages of the lower number of transitions per
period.
The implementation of the 12-sector patterns on the DSP
microcontroller is challenging, since each switch might have
more than one transition per period. However, the imple-
mentation of the 24-sector ones is simpler, because there is
maximum one transition per period.
Fig. 10 shows the THD of the output line voltage of a
six-phase inverter using different switching patterns at the
same switching frequency at different modulation indexes.
The differences are not very high in between; however,
the 24-sector continuous modulation switching pattern shows
the lowest profile of THD. Based on the comparisons, the pat-
terns based on 24-sector modulation prove to be the best
options for modulating six-phase inverters in traction appli-
cations, as the switching losses will be lower, which results in
lower cooling requirements and higher efficiency operation to
achieve extended-range vehicles.
Vector Classification Technique (VCT): This technique is
mainly based on the double-DQ theory of modeling the six-
phase system. VCT was proposed in [70] and followed by

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS 1337

TABLE IV
P OSSIBLE S WITCHING PATTERNS OF SVPWM T ECHNIQUE FOR A SYMMETRICAL S IX -P HASE BASED ON VSD T HEORY

Fig. 10. Comparison between different switching patterns in line voltage

THD at 3150-Hz switching frequency.

further enhancements in choosing the vectors [79] using a

modified neural network and getting better utilization of the
dc-link voltage, such as in [80]. Two different approaches have
approached the realization of this technique as in Fig. 11. The
first way is implemented by using two references rotating in
the familiar hexagon, representing all the possible switching
vectors of the three-phase inverter, as shown in Fig. 11(a).
The second way is by rotating one of the hexagons by 30◦
and using a single reference, such as in Fig. 11(b). A study
proposed in [81] compares the classification technique and the
12-sector switching patterns. The study shows that the clas-
sification technique is simpler to implement and gives higher
utilization of the dc-link voltage, but the SVPWM techniques
based on VSD theory gives better harmonic minimization.
A comparison between the modulation techniques of
six-phase inverters is shown in Table V. The carrier-based
modulation techniques are simple to implement and can get to Fig. 11. Vector classification technique realized by (a) two separate references
and (b) a single reference [82].
high modulation indexes in the linear region of modulation by
injecting zero-sequence harmonics to the references. However,
the SVPWM methods are superior in terms of the ability method is more complex to implement, but using the switching
to enhance the quality of the output waveform using the patterns mentioned in the previous subsections can result in a
redundant switching states. The SVPWM based on the VSD significant reduction in the xy currents.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 1338 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019
TABLE V
M ODULATION S CHEMES OF S IX -P HASE , T WO -L EVEL I NVERTERS
Fig. 12. Nine-switch inverter topology.
B. Reduced Switch-Count Inverters for MPDs
Increasing the power density of power electronics converters
is one of the trends in EV applications [83]. One of the
solutions to reduce the size of the converter is using reduced
switch-count topologies. The Nine-Switch Inverter (NSI) is
one of the topologies that has been investigated in the literature
as a six-phase inverter for multiphase machines [4], [84], [85].
Fig. 12 shows the configuration of the inverter. The modulation
techniques are discussed in [86]–[93], which show that the
maximum modulation index for supplying an asymmetrical
six-phase machine is around 81%, which is one of the disad-
vantages of this converter topology. The switching frequency
of the middle switches is double the switching of the other
switches, i.e., the switching losses are the same as the con-
ventional 12-switch topology. Several control techniques can
be used with the topology, as stated in [84] and [94]–[96].
The Z-source NSI topology comprises Hybrid EV (HEV)
applications in [97] and [98].
Five-leg inverter is another topology with a smaller number
of switches that was introduced as an alternative to the
conventional inverter for six-phase drives [99], as shown in
Fig. 13. The topology is based on connecting one machine
phase to the midpoint of the dc-link as a reference to the
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
Fig. 13. Five-leg inverter topology.
other phases. The main challenge of using such a topology is
the balancing of the midpoint voltage [100].
C. Dual Supply Inverters for Open-End Winding Machines
Dual supply inverter topology is a converter topology used
to supply machines with an open-end winding configuration.
This topology is supplied from isolated dc sources, so it
does not require a capacitor voltage balancing technique,
unlike most of the multilevel inverter topologies. The cascaded
H-Bridge multilevel topology is also supplied from isolated
dc sources, which are typically provided by custom-designed
phase-shifting transformers. In EV applications, the dc source
is typically a pack of batteries which can be rearranged and
configured to supply such topologies. Lower dc-link voltages
are one of the merits of using such converters with an increased
number of components.
Several topologies were introduced in the literature based
on the supplies used and how to connect them. One of them is
the topology with two-sided supplies to supply two six-phase
inverters, and each phase winding is connected to a converter
output from both phase terminals, and it will be referred to in
this article as dual-supply topology I as in Fig. 14. Another
topology uses only one supply (dual supply topology II) to
supply the two inverters in a five-phase drive to eliminate the
common-mode voltage by means of SVPWM [42].
Azeez et al. [101] discusses a different topology with
four three-phase inverters and two supplies connected from
both sides with a capacitor commonly shared by the middle
inverters (dual supply topology III), as shown in Fig. 15.
This topology is used to suppress the fifth and seventh har-
monics from the output voltage at the cost of adding one more
capacitor bank. The assessment of the converter performance
is discussed in [102]–[104]. The analysis of the output current
ripples in the drive is investigated in [102], regarding the
importance of minimizing such ripples for better performance
of the drive. The converter can generate a three-level output,
which is better than the conventional two-level one in terms
of the resulting ripples in the output currents.
A comparison between the performance of symmetrical and
asymmetrical six-phase drives with an open-end configuration
is discussed in [104] using different modulation techniques.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS 1339

Fig. 14. Dual-supply, inverter-fed, open-end, winding six-phase machine (topology I).

Fig. 15. Dual-supply, inverter-fed, open-end, winding six-phase machine based on three-phase inverters (topology III).

The comparison shows that the symmetrical six-phase drives

have the same results as the three- and five-phase drives,
while the asymmetrical ones behave differently according to
the different techniques applied.

D. Multilevel Inverters for MPDs

Multilevel inverters are a famous solution for high-voltage,
high-power motor drives. Several topologies are discussed in
the literature with pros and cons based on their cost, size,
performance, and complexity. The complexity is increased in
multiphase, multilevel drives due to the increased number of
switches and switching states. However, since there is a push
to increase the dc-link voltage (>800 V) in the next generation
of EVs [67], multilevel topologies will be a good candidate
for improving the output power quality, while improving the
overall efficiency with the low-voltage power semiconductors.
One of the well known and used multilevel topologies is the
neutral-point clamped inverter (NPC). The research up until
this article is focused on the three-level NPC inverter to supply
multiphase machines due to the complexity of the system.
Fig. 16 shows a three-level NPC-fed, six-phase machine.
Similar to the two-level topology, the SVPWM techniques
used for multilevel multiphase drives (MPDs) are classification
based and VSD based. The vector classification technique
alongside the modeling of the drive is investigated in [105].
VSD-based models are also discussed in [65] and [106],
the latter publication adding a decoupling, especially for
developing a switching pattern based on VSD. The num-
ber of switching states to supply dual three-phase machines
is 729 states with 665 distinctive voltage vectors. It is
quite the task to choose the optimum switching vector to
ensure the balancing of the midpoint of the dc-link and
the minimal output circulating currents. Wang et al. [106]
Fig. 16. NPC-fed, six-phase machine.
demonstrated a switching pattern for operating the converter,
taking into account the elimination of the xy voltage com-
ponents. An improved approach is introduced in [65] which
proposes a decoupled modulation technique to address the
harmonic components from the back-EMF generated due to
unbalanced parameters of the machine and to eliminate the
zero-sequence components. The improved technique has a
major drawback represented in the complexity and higher
switching frequency.
A T-NPC topology for MPDs has also been discussed in
the literature and the modulation technique of the topology
is investigated in [107], exploiting the voltage classifica-
tion technique. The Direct Torque Control (DTC) control
method is applied in [108] to control the drive, which is
also based on the double SVPWM. A topology based on the
CHB structure to supply a symmetrical six-phase machine is
discussed in [109] as a medium-voltage drive. As an extension

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 1340 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019
to dual-supply inverter topology, the two-level, three-phase
topology is replaced with the three-level, three-phase NPC,
as discussed in [110]. Melo et al. [111] proposed a hybrid
inverter to supply one three-phase set from a conventional
inverter and the other set from an NPC inverter. Some of
the advanced multilevel topologies based on FC, CHB, and
active neutral-point clamped inverter (ANPC) to provide a high
number of output levels up to 49 are introduced in [112]–[114].
Almost all the multilevel topologies for MPDs have been
developed for medium-voltage industrial motor drives, and to
the best of author’s knowledge, no such multilevel topologies
have been used for EV applications so far. A summary
comparison between the aforementioned converter topologies
is illustrated in Table VI.
V. C ONTROL M ETHODS
The control techniques used to obtain the optimal perfor-
mance of MPDs are the current control method, the classical
FOC, and the recent model-based schemes such as MPC.
All these schemes and their applications are discussed in this
section.
A. Harmonic Current Mitigation Methods
As discussed earlier, the currents in the xy subspace produce
losses and have no contribution to the torque production
process. These currents are produced because of the voltage
harmonics resulting from modulation technique or deadtime.
The resulting current harmonics have a measurable amplitude
even if the voltage harmonics are very small due to the
small equivalent impedance of the xy subspace. The solutions
proposed in the literature use controllers to nullify these
currents by applying a reference of zero-value, and the mission
of the controllers is to track the reference. The application of
PI controllers in the synchronous frame for several multiphase
induction machines was investigated in [115].
The case of asymmetrical six-phase induction motors is
addressed in [29], which proposed applying dual PI controllers
to the synchronous and antisynchronous frames to mitigate
the unwanted harmonics in the xy subspace and applying
resonant controllers to compensate for the inserted deadtime.
The application of resonant controllers in the xy subspace
for the same machine was addressed in [116]. The idea of
applying multiresonant control in the synchronous frames
has been proposed in [117] to compensate current harmon-
ics in symmetrical multiphase machines. The asymmetrical,
six-phase PMSMs were investigated in [118] to compensate
harmonics using a disturbance observer based on the double dq
machine model. The compensation based on the multiresonant
controllers and the VSD machine model is studied in [119].
The efficiency of many MPDs including inverter losses for
light load operating conditions has been studied in [120].
The complexity and increased number of controllers lead to
a stability issue for industrial applications such as EVs.
B. FOC
FOC is one of the famous vector control schemes used
in electric drives to achieve the desired torque-speed char-
acteristics with good transient performance. The main key to
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
Fig. 17. Block diagram of applying MPC to six-phase drives.
apply FOC is to apply the rotor angle for decomposition of
the current components into the torque and flux controlling
components. The rotor flux angle is estimated in the induction
machine using observers and estimators to account for the
slip between the rotor speed and the flux speed. For PMSM,
there is no slip calculation, and the measured rotor angle can
be applied directly to find the dq axes’ current components
and control them to the desired characteristics. The difference
between applying FOC in three-phase drives and six-phase
drives is that multiphase machines have extra subspaces, which
represent losses as the xy subspace. The control scheme should
have current controllers to eliminate these currents. Sensorless
control of five-phase machines is also discussed in [122] and
a comparison between using PI controllers and fuzzy logic
controllers is investigated in [123].
C. MPC
The MPC control method was introduced in [124] to
control power converters, generally providing a fast dynamic
response. The concept of applying such a method to control
asymmetrical six-phase drives is discussed in [125]. The
concept is that the switching pattern for each sampling period
is determined based on optimizing a cost function that uses the
model of the drive system in the discrete-time domain, shown
in Fig. 17. All possible switching patterns are applied to the
cost function to determine which one is the best candidate
to be applied. The cost function can be used to satisfy the
requirements of the drive, such as meeting a certain THD
of the output waveform, eliminating circulating currents of
the xy subplane, providing fault-tolerance control [126]–[128],
control of the machine torque, and even mixing some criteria
in one cost function, but with different weights according to
their priority level.
An interesting addition to the control method is that instead
of applying one voltage vector in each sampling period, one
can apply two of them with different weights to achieve the
desired output in the fundamental plane and a zero-average
voltage in the xy plane as well. This method is called the
MPC using virtual voltage vectors, that have been introduced
in [129]. The only disadvantage to applying this method
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS 1341

TABLE VI
C OMPARISON B ETWEEN D IFFERENT C ONVERTER T OPOLOGIES FOR S IX -P HASE D RIVES

compared to the former finite-control set method is the TABLE VII

decreased utilization of the dc-link voltage by 7.2%. The C OMPARISON B ETWEEN D IFFERENT C ONTROL M ETHODS
FOR S IX -P HASE D RIVES
main problems regarding applying MPC is the need to exactly
determine the parameters of the drive and the increased time
of the sampling period as a large number of computations
is needed. A simplified MPC method is developed in [130]
to reduce the computational time in the microprocessor to
make the method more feasible and practical. Table VII shows
a comparison between FOC and MPC methods to control a
six-phase drive. The FOC method has been used for decades
in industrial motor drives; however, the MPC method is a
promising method compared to FOC, since it is simple to
implement, the modulation scheme is already included in the
control method, there is no need to tune the parameters of the
controllers and the transient response is fast compared to FOC.
The MPC method is considered in this article as a trend in the
control area and its challenges are covered in Section VI.

VI. C URRENT S TATUS IN U SE OF MPD S

IN T RACTION A PPLICATIONS
Recently, TM4 launched a manufacturing line of MPDs
using three-, six-, and nine-phase machines. Fig. 18 shows
one of the products in the line called TM4 SUMOTM4 MD.
The product has several models with a peak power range
between 162 and 265 kW and a maximum torque ranging from
1600 to 3255 Nm. The product series is suited for medium to
heavy-duty vehicles such as buses and delivery trucks. One
of the EV manufacturers that uses the TM4 products is the
EMOSS Company in their EMS truck series with a six-phase
model up to 200 kW/2950 Nm and a nine-phase model up
to 250 kW/3400 Nm. The torque-speed characteristics of
different models of the series are shown in Fig. 19 and the
six-phase models have higher curves compared with the three-
phase ones.
TM4 also provides another series called SUMOTM4 HD
series dedicated for heavy-duty applications. This series uses
nine-phase machines with their high voltage inverter. This
series models provide high characteristics such as peak power
up to 350 kW, peak torque up to 3500, and maximum operating
speed up to 3400 RPM. This series is integrated with Dana’s
products for heavy-duty applications like buses and trucks.
Another example of empowering the use of multiphase
machines in the industry is studied in [131]. The proposal
is to study the use of a nine-phase traction motor in an ultra-
high-speed elevator as an alternative to traditional three-phase
PMSM. The application requires a high-power motor to

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
 1342 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019
Fig. 18. SUMOTM4 MD six-phase drive [121].
achieve speeds such as 1000 m/min and carry a weight
of 1600 Kg which is equivalent to 24 passengers to travel
a distance of 540 m. The estimated power of the motor is
1 MW and the nine-phase alternative provides the opportunity
to divide this high-power demand across a higher number of
phases. The experimental tests were carried out in one of the
world’s tallest test towers as this article claims, and the results
show that the proposed system is reliable and can operate to
achieve the required standards. Due to the superior features
of MPDs in terms of power density, performance, and fault-
tolerant capability, it is expected to see more manufacturers
produce MPDs for EV, ship, and electric aircraft applications.
VII. R ESEARCH T RENDS AND C HALLENGES IN U SE OF
MPD S IN T RACTION A PPLICATIONS
A. Topological Advances
1) New Topologies: The established inverter topologies for
MPDs require an additional number of switches compared to
the conventional three-phase inverters to control the inverter
properly and provide the given features discussed in this
article. This will be more complex and expensive where the dc-
link voltage is higher than 800 V and multilevel topologies will
be very attractive to improve the torque and THD performance
while using the low-voltage power semiconductor devices.
It seems that multilevel topologies with a reduced switch count
are an interesting area of research to employ MPDs in the next
generation of high-voltage dc-link EVs and take the benefits
of high-torque density and fault-tolerant operation.
2) EVs With a Highly Integrated Modular Drivetrain:
The idea of the integration of an inverter with the machine
inside the drivetrain is to minimize the connections from the
inverter to the machine phases to overcome the disadvantage
of increased number of phases and thus the connecting wires.
An asymmetrical nine-phase machine with 20◦ phase shift
between the three groups of three-phase windings with a
highly integrated nine-phase drivetrain is shown in Fig. 20.
The drivetrain developed in [132] is a 60 kW with a 35 kW/L
power density of the electronics part as a powerful solution
to the sizing of the drive inside the vehicle. The maximum
drive speed is 11 500 rpm with a peak torque of 170 Nm.
The benefits reported in [132] are the easy integration of the
drive, saving space in a limited environment such as EVs, and
low currents; since the drive is a nine-phase one, the converter
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
Fig. 19. Torque-speed characteristics of different models of SUMOTM4 MD
series [121].
Fig. 20. Highly integrated nine-phase drivetrain for EV applications.
(1) Motor housing. (2) Stator. (3) Ferrite motor. (4) Bearing shield-inverter
base plate. (5) Inverter [132].
can be designed using lower class modules, which are of low
price and low size.
3) EVs With Highly Integrated Drivetrain: The idea of the
integration of an inverter with the machine inside the drivetrain
is to minimize the connections from the inverter to the machine
phases to overcome the disadvantage of the increased number
of phases, and thus the connecting wires.
An asymmetrical nine-phase machine with 20◦ phase shift
between the three groups of three-phase windings with a
highly integrated nine-phase drivetrain is shown in Fig. 20.
The drivetrain developed in [132] is a 60 kW with a 35 kW/L
power density of the electronic part as a powerful solution
to the sizing of the drive inside the vehicle. The maximum
drive speed is 11 500 rpm with a peak torque of 170 Nm. The
benefits reported in [132] are the easy integration of the drive,
saving space in limited environments such as EVs, and low
currents; since the drive is a nine-phase one the converter can
be designed using lower class modules which are of low price
and low size.
The challenges to such a topology can be in the ventilation,
thermal management, and cooling from a mechanical point of
view. From an electrical point of view, the high electromag-
netic interference (EMI) and its associated problems must be
studied thoroughly since the inverter and its driving circuits
are very close to the motor in this design.
Using an interleaving strategy in [133] could lead to a
reduction in the dc-link current ripples and the size of the
dc-link capacitors. A case study is presented in [133], which
considers a 15-phase integrated modular drive, and the reduc-
tion in the capacitor sizing is up to half of its original
size. In [134], a general solution to the design of dc-link
capacitors of integrated modular drives is introduced based on
a multicriteria, multidegrees of freedom optimization problem.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS
Fig. 21. Fast, onboard, battery charger using a six-phase drive [136].
4) Integrated On-Board Battery Chargers: An emerging
research topic in EV applications is the topic of using the
on-board motor and power electronics converters as chargers
to the batteries of the vehicle to avoid the installation of
additional converters in the off-board chargers and use just
the terminals provided by the grid. The options are to connect
a single phase for slow charging to the vehicle or the three
phases for fast charging. The challenge is to just use the
available gear to accomplish the charging process and at
the same time prevent the machine from revolving. Stopping
the movement can be achieved by either mechanical locks or
special connections in the electrical power circuit. Mechanical
locks are an issue because of the size and weight besides
the extra electrical losses in the machine windings; however,
the goal is to generate a pulsating field in the fast charging
mode to achieve the charging without any movement.
For asymmetrical six-phase drives, the fast charging is
implemented by two configurations. One of the configurations
was introduced in [135], which uses a transformer with two
secondary sets to provide a six-phase supply along with a
phase transposition. The drawbacks of the configuration are
that the custom-made transformer is costly and adding to the
overall size.
The study in [136] discusses the other configuration in
which the three-phase grid terminals are connected to the six-
phase inverter, as each grid phase is connected to two phases
of the inverter. Fig. 21 shows the connection and the current
flow in different phases. It is important to mention that the
connections of the phase windings are altered by means of
contactors to change from the propulsion mode (two separate
neutral connections) to the charging mode (special connection
to the three-phase grid). For single-phase grid connection, the
problem of rotation disappears because the field in the machine
is pulsating and the two neutral points are connected to grid
terminals, as shown in Fig. 22. The charging current is the
zero-sequence component [137] and nullifying other subplanes
to stop revolving while minimizing the losses.
The control of former topologies depends on the mode of
operation. The control technique is well established in the
propulsion mode and no modification is required. The charging
mode aims to control the currents in the machine windings
to just charge the battery and produce a zero-average torque.
For fast charging, the connection of the phases to the grid
terminals ensures the production of a pulsating field, so the
system is equivalent to a three-phase rectifier connected to
a grid. A voltage-oriented control is adopted with a minor
modification of adding extra controllers in the antisynchronous
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
1343
Fig. 22. Single-phase, onboard, battery charger using a six-phase drive [137].
frame to eliminate the asymmetries in the phase currents
resulting from different equivalent impedances of the machine.
The vehicle to grid mode is the same as the charging mode
with the difference in the opposite power flow direction.
Although the integrated on-board battery charger idea has
been developed for a six-phase drive, there are still some
challenges that require more attention by researchers, such as
lowering the THD of the grid currents to meet the standards,
substituting the auxiliary contactors to change the connection
of the windings, and integrating the PWM rectifier and the
dc–dc converter together.
B. Cost of Manufacturing MPDs
One of the main challenges that faces a new technology such
as MPD and limits the spread of it is the availability of the
well-established three-phase drives. The cost of producing new
technologies is high in the beginning because the traditional
drives are off the shelf and differences are not high enough as a
breakthrough technology to force the manufacturers to produce
more MPDs. The advances in the industrial power electronics
and the fact that transportation applications depend on dc–ac
converters make MPDs more likely to be used in the future
as the difference would be inverting from dc to any number
of phases as long as the drive would be more efficient on the
overall performance basis.
According to [138], the U.S Department of Energy (DOE)
targets to reduce the cost of producing electric traction motors
by 30% and increase the power density by 80% by the
year 2025 compared to the targets of the year 2020 as
in Table VIII. The features of MPDs such as the reduction
of the required dc-link capacitance and the torque density
improvement could be a viable solution to close these targets
and achieve better drives for EV applications. The comparison
between three-phase and six-phase drives shows from the
aforementioned subsections that the reduction in the dc-link
bank is about half and the use of the harmonic injection
method could be used to scale-up smaller drives in weight
and size.
C. Control Methods
The model-based control techniques such as MPC methods
have shown better performance at the expense of complexity
and a huge number of computations. This is getting worse
when the number of phases increases. The challenge facing
 1344 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019
TABLE VIII
DOE TARGETS FOR E LECTRIC T RACTION M OTORS [138]
the development of these control techniques for MPDs is the
feasibility of implementing these controllers in the current
DSP microcontrollers, matching the speed and memory limits.
Therefore, further reduction in the number of computations
while retaining the best performance of MPDs in EVs is a big
challenge.
These advanced control techniques need to be developed to
reduce the dc-link ripple currents and thus reduce the dc-link
capacitor values which lead to further improvement of the
power density.
D. Use of Wide Bandgap Devices in MPDs
The technology of wide bandgap devices is the next gen-
eration in the power electronics industry that will enable
manufacturing high-power electric drives largely. Silicon car-
bide (SiC) and gallium nitride (GaN) are the most used
materials to produce WBG devices. The WBG modules are
better than conventional silicon (Si) modules in terms of
losses, weight breakdown voltage, switching frequency, and
withstanding higher junction temperatures [139]. A case study
is investigated in [140] to design a 1200-V, 100-A, VSC with
SiC-based modules. The study includes a comparison between
SiC modules and Si modules at different operating switching
frequencies in terms of the estimated losses. The comparison
shows the prevalence of SiC modules at all the frequencies and
also shows that SiC can even operate at 50 kHz at which Si
modules cannot. In [141]–[143], the focus is on the application
of WBG devices in EV applications. The problem of dv/dt is
addressed in [144] and a proposed solution is included using
a multistage drive circuit.
The lower losses associated with using WBG devices incites
increasing the switching frequency to obtain better output
waveforms, especially in low-inductance drive applications.
The challenge of high dv/dt arises with the application of WBG
devices accompanied by using a high switching frequency
at high voltages. Another challenge that faces the spread of
industrial manufacturing of WBG devices is the cost. The
cost is expected to decline if the manufacturing technologies
advance and the market demands change.
VIII. C ONCLUSION
This article discussed the remarkable features of MPDs that
qualify them as a competitive solution in EV applications
such as using lower rating semiconductor devices, operation
under faulty conditions with optimal current control, ability to
increase the torque density of the drive via harmonic injection,
and smaller dc-link size. This article also presented the recent
advances in several power electronics topology to implement
in the drive and their modulation techniques. From the features
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
given for MPDs, it can be concluded that this structure can be
an interesting solution for the next generation of EVs.
R EFERENCES
[1] E. E. Ward and H. Härer, “Preliminary investigation of an invertor-
fed 5-phase induction motor,” Proc. Inst. Elect. Eng., vol. 116, no. 6,
pp. 980–984, 1969.
[2] M. A. Abbas, R. Christen, and T. M. Jahns, “Six-phase voltage source
inverter driven induction motor,” IEEE Trans. Ind. Appl., vol. IA-20,
no. 5, pp. 1251–1259, Sep. 1984.
[3] N. Bodo, E. Levi, I. Subotic, J. Espina, L. Empringham, and
C. M. Johnson, “Efficiency evaluation of fully integrated on-board
EV battery chargers with nine-phase machines,” IEEE Trans. Energy
Convers., vol. 32, no. 1, pp. 257–266, Mar. 2017.
[4] M. S. Diab, A. A. Elserougi, A. S. Abdel-Khalik, A. M. Massoud, and
S. Ahmed, “A nine-switch-converter-based integrated motor drive and
battery charger system for EVs using symmetrical six-phase machines,”
IEEE Trans. Ind. Electron., vol. 63, no. 9, pp. 5326–5335, Sep. 2016.
[5] I. Subotic, N. Bodo, E. Levi, M. Jones, and V. Levi, “Isolated chargers
for EVs incorporating six-phase machines,” IEEE Trans. Ind. Electron.,
vol. 63, no. 1, pp. 653–664, Jan. 2016.
[6] V. F. Pires, A. Cordeiro, D. Foito, and J. F. Silva, “A three-phase on-
board integrated battery charger for EVs with six-phase machine and
nine switch converter,” in Proc. IEEE 13th Int. Conf. Compat., Power
Electron. Power Eng. (CPE-POWERENG), Apr. 2019, pp. 1–6.
[7] H. Zahr, F. Scuiller, and E. Semail, “Five-phase SPM machine with
electronic pole changing effect for marine propulsion,” in Proc. Int.
Conf. Elect. Syst. Aircr., Railway, Ship Propuls. Road Vehicles Int.
Transp. Electrific. Conf. (ESARS-ITEC), Nov. 2016, pp. 1–6.
[8] M. Qiao, C. Jiang, y. Zhu, and G. Li, “Research on design method
and electromagnetic vibration of six-phase fractional-slot concentrated-
winding PM motor suitable for ship propulsion,” IEEE Access, vol. 4,
pp. 8535–8543, 2016.
[9] E. Levi, “Multiphase electric machines for variable-speed applications,”
IEEE Trans. Ind. Electron., vol. 55, no. 5, pp. 1893–1909, May 2008.
[10] Z. Liu, Y. Li, and Z. Zheng, “A review of drive techniques for
multiphase machines,” CES Trans. Elect. Mach. Syst., vol. 2, no. 2,
pp. 243–251, Jun. 2018.
[11] E. Levi, N. Bodo, O. Dordevic, and M. Jones, “Recent advances
in power electronic converter control for multiphase drive systems,”
in Proc. IEEE Workshop Elect. Mach. Design, Control Diagnosis
(WEMDCD), Mar. 2013, pp. 158–167.
[12] M. J. Duran and F. Barrero, “Recent advances in the design, model-
ing, and control of multiphase machines—Part II,” IEEE Trans. Ind.
Electron., vol. 63, no. 1, pp. 459–468, Jan. 2016.
[13] F. Barrero and M. J. Duran, “Recent advances in the design, model-
ing, and control of multiphase machines—Part I,” IEEE Trans. Ind.
Electron., vol. 63, no. 1, pp. 449–458, Jan. 2016.
[14] F. Baneira, J. Doval-Gandoy, A. G. Yepes, Ó. López, and
D. Pérez-Estévez, “Control strategy for dual three-phase PMSMs with
minimum losses in the full torque operation range under single open-
phase fault,” in Proc. IEEE Energy Convers. Congr. Expo. (ECCE),
Sep. 2016, pp. 1–8.
[15] W. N. W. A. Munim, M. J. Duran, H. S. Che, M. Bermúdez,
I. González-Prieto, and N. A. Rahim, “A unified analysis of the fault
tolerance capability in six-phase induction motor drives,” IEEE Trans.
Power Electron., vol. 32, no. 10, pp. 7824–7836, Oct. 2017.
[16] X. Wang, Z. Wang, Z. Xu, M. Cheng, W. Wang, and Y. Hu, “Com-
prehensive diagnosis and tolerance strategies for electrical faults and
sensor faults in dual three-phase PMSM drives,” IEEE Trans. Power
Electron., vol. 34, no. 7, pp. 6669–6684, Jul. 2019.
[17] H. S. Che, M. J. Duran, E. Levi, M. Jones, W.-P. Hew, and
N. A. Rahim, “Postfault operation of an asymmetrical six-phase
induction machine with single and two isolated neutral points,” IEEE
Trans. Power Electron., vol. 29, no. 10, pp. 5406–5416, Oct. 2014.
[18] A. G. Yepes, J. Doval-Gandoy, F. Baneira, and H. Toliyat, “Postfault
strategy for dual three-phase machines with minimum loss in the full
torque operation range under two open phases,” in Proc. IEEE Energy
Convers. Congr. Expo. (ECCE), Sep. 2018, pp. 3380–3385.
[19] J.-R. Fu and T. A. Lipo, “Disturbance free operation of a multiphase
current regulated motor drive with an opened phase,” in Proc. 28th IAS
Annu. Meeting Conf. Rec. IEEE Ind. Appl. Conf., vol. 1, Oct. 1993,
pp. 637–644.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS
[20] R. O. C. Lyra and T. A. Lipo, “Torque density improvement in a six-
phase induction motor with third harmonic current injection,” in Proc.
36th IAS Annu. Meeting Conf. Rec. IEEE Ind. Appl. Conf., vol. 3,
Sep./Oct. 2001, pp. 1779–1786.
[21] K. Wang, Z. Q. Zhu, Y. Ren, and G. Ombach, “Torque improve-
ment of dual three-phase permanent-magnet machine with third-
harmonic current injection,” IEEE Trans. Ind. Electron., vol. 62, no. 11,
pp. 6833–6844, Nov. 2015.
[22] Y. Hu, Z. Q. Zhu, and M. Odavic, “Torque capability enhancement of
dual three-phase PMSM drive with fifth and seventh current harmonics
injection,” IEEE Trans. Ind. Appl., vol. 53, no. 5, pp. 4526–4535,
Sep./Oct. 2017.
[23] G. Feng, C. Lai, M. Kelly, and N. C. Kar, “Dual three-phase PMSM
torque modeling and maximum torque per peak current control through
optimized harmonic current injection,” IEEE Trans. Ind. Electron.,
vol. 66, no. 5, pp. 3356–3368, May 2019.
[24] F. B. Bendixen, F. Blaabjerg, P. O. Rasmussen, P. Vadstrup, and
K. Krabbe, “Controlling the DC-link midpoint potential in a six-phase
motor-drive,” in Proc. IEEE 35th Annu. Power Electron. Spec. Conf.,
vol. 3, Jun. 2004, pp. 2128–2132.
[25] A. S. Abdel-Khalik, A. M. Massoud, and S. Ahmed, “Effect of DC-link
voltage limitation on postfault steady-state performance of asymmetri-
cal six-phase induction machines,” IEEE Trans. Ind. Electron., vol. 65,
no. 9, pp. 6890–6900, Sep. 2018.
[26] A. S. Abdel-Khalik, M. I. Masoud, and B. W. Williams, “Improved
flux pattern with third harmonic injection for multiphase induction
machines,” IEEE Trans. Power Electron., vol. 27, no. 3, pp. 1563–1578,
Mar. 2012.
[27] Y. Zhao and T. A. Lipo, “Space vector PWM control of dual three-phase
induction machine using vector space decomposition,” IEEE Trans. Ind.
Appl., vol. 31, no. 5, pp. 1100–1109, Sep./Oct. 1995.
[28] Y. Zhao and T. A. Lipo, “Modeling and control of a multi-phase
induction machine with structural unbalance,” IEEE Trans. Energy
Convers., vol. 11, no. 3, pp. 570–577, Sep. 1996.
[29] H. S. Che, E. Levi, M. Jones, W.-P. Hew, and N. A. Rahim, “Cur-
rent control methods for an asymmetrical six-phase induction motor
drive,” IEEE Trans. Power Electron., vol. 29, no. 1, pp. 407–417,
Jan. 2014.
[30] Manjesh cp and A. S. Ananda, “Analysis and study of total harmonic
distortion in five phase PWM inverter using LC filter with neutral point
DC link capacitor voltage balancing scheme,” in Proc. Annu. Int. Conf.
Emerg. Res. Areas, Magn., Mach. Drives (AICERA/iCMMD), Jul. 2014,
pp. 1–4.
[31] D. Dujic, M. Jones, and E. Levi, “Continuous carrier-based vs. space
vector PWM for five-phase VSI,” in Proc. Int. Conf. ‘Comput. Tool’
(EUROCON), Sep. 2007, pp. 1772–1779.
[32] Deni, E. G. Supriatna, and P. A. Dahono, “Output current ripple
analysis of five-phase PWM inverters,” in Proc. 7th Int. Conf. Power
Electron. Drive Syst., 2007, pp. 1074–1080.
[33] K. A. Chinmaya and G. K. Singh, “Modeling and comparison of
space vector PWM schemes for a five-phase induction motor drive,” in
Proc. 44th Annu. Conf. IEEE Ind. Electron. Soc. (IECON), Oct. 2018,
pp. 559–564.
[34] Y. N. Tatte and M. V. Aware, “Torque ripple and harmonic current
reduction in a three-level inverter-fed direct-torque-controlled five-
phase induction motor,” IEEE Trans. Ind. Electron., vol. 64, no. 7,
pp. 5265–5275, Jul. 2017.
[35] S. Payami, R. K. Behera, A. Iqbal, and R. A. Al-Ammari,
“Common-mode voltage and vibration mitigation of a five-phase
three-level NPC inverter-fed induction motor drive system,” IEEE
J. Emerg. Sel. Topics Power Electron., vol. 3, no. 2, pp. 349–361,
Jun. 2015.
[36] S. Lihitkar and S. Rangari, “Comparative study of IPD and POD for
three level five phase inverter,” in Proc. Int. Conf. Smart Electr. Drives
Power Syst. (ICSEDPS), Jun. 2018, pp. 313–316.
[37] Y. N. Tatte and M. V. Aware, “Torque ripple minimization in five-
phase three-level inverter fed direct torque control induction motor
drive,” in Proc. 17th Eur. Conf. Power Electron. Appl. (EPE ECCE-
Eur.), Sep. 2015, pp. 1–6.
[38] B. Sakthisudhursun, J. K. Pandit, and M. V. Aware, “Simplified three-
level five-phase SVPWM,” IEEE Trans. Power Electron., vol. 31, no. 3,
pp. 2429–2436, Mar. 2016.
[39] I. N. W. Satiawan, I. B. F. Citarsa, and I. K. Wiryajati, “A simple PWM
scheme for a four-level dual-inverter fed open-end winding five-phase
motor drive,” in Proc. 4th IET Clean Energy Technol. Conf. (CEAT),
Nov. 2016, pp. 1–7.
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
1345
[40] M. Priestley, J. E. Fletcher, and C. Tan, “Space-vector PWM tech-
nique for five-phase open-end winding PMSM drive operating in the
overmodulation region,” IEEE Trans. Ind. Electron., vol. 65, no. 9,
pp. 6816–6827, Sep. 2018.
[41] M. Darijevic, N. Bodo, and M. Jones, “On the five-phase open-
end winding drives performance,” in Proc. PCIM Eur., Int. Exhib.
Conf. Power Electron., Intell. Motion, Renew. Energy Energy Manage.,
May 2015, pp. 1–8.
[42] N. Bodo, M. Jones, and E. Levi, “A space vector PWM with common-
mode voltage elimination for open-end winding five-phase drives with
a single DC supply,” IEEE Trans. Ind. Electron., vol. 61, no. 5,
pp. 2197–2207, May 2014.
[43] S. Belkhode, R. Karampuri, and S. Jain, “Modified phase shifted
SVPWM technique for five-phase open-end winding induction motor
with zero common mode voltage,” in Proc. 14th IEEE India Council
Int. Conf. (INDICON), Dec. 2017, pp. 1–6.
[44] D. Zhang, B. Xu, H. Yang, and P. Zhu, “Simulation analysis of
SVPWM based on seven-phase permanent magnet synchronous motor,”
in Proc. Int. Conf. Control, Automat. Inf. Sci. (ICCAIS), Oct./Nov. 2017,
pp. 251–256.
[45] J. Prieto, J. A. Riveros, and M. Rivera, “Development of synchronized
voltage modulation techniques for seven-phase voltage source invert-
ers,” in Proc. IEEE 12th Int. Conf. Compat., Power Electron. Power
Eng. (CPE-POWERENG), Apr. 2018, pp. 1–6.
[46] M. A. Khan, S. K. M. Ahmed, A. Iqbal, H. A. Rub, and
S. K. Moinoddin, “Discontinuous space vector PWM strategies for a
seven-phase voltage source inverter,” in Proc. 35th Annu. Conf. IEEE
Ind. Electron., Nov. 2009, pp. 397–402.
[47] D. Dujic, E. Levi, M. Jones, G. Grandi, G. Serra, and A. Tani,
“Continuous PWM techniques for sinusoidal voltage generation with
seven-phase voltage source inverters,” in Proc. IEEE Power Electron.
Spec. Conf., Jun. 2007, pp. 47–52.
[48] O. Dordevic, E. Levi, and M. Jones, “A vector space decomposition
based space vector PWM algorithm for a three-level seven-phase
voltage source inverter,” IEEE Trans. Power Electron., vol. 28, no. 2,
pp. 637–649, Feb. 2013.
[49] N. Bodo, M. Jones, and E. Levi, “Multi-level space-vector PWM
algorithm for seven-phase open-end winding drives,” in Proc. IEEE
Int. Symp. Ind. Electron., Jun. 2011, pp. 1881–1886.
[50] Z. Pan and J. Ikaheimo, “Control optimization of nine-phase variable
speed drive,” in Proc. IEEE Int. Electr. Mach. Drives Conf. (IEMDC),
May 2015, pp. 910–915.
[51] Z. Pan and J. Ikaheimo, “Nine-phase variable speed drive system with
reduced switching PWM,” in Proc. Int. Conf. Electr. Mach. (ICEM),
Sep. 2014, pp. 823–828.
[52] V. F. M. B. Melo, C. B. Jacobina, and N. B. de Freitas, “Open-end nine-
phase machine conversion systems,” IEEE Trans. Ind. Appl., vol. 53,
no. 3, pp. 2329–2341, May/Jun. 2017.
[53] V. F. M. B. Melo, C. B. Jacobina, and N. B. de Freitas, “Open-end nine-
phase machine conversion systems,” in Proc. IEEE Energy Convers.
Congr. Expo. (ECCE), Sep. 2015, pp. 6934–6941.
[54] D. Dujic, M. Jones, and E. Levi, “Space vector PWM for nine-
phase VSI with sinusoidal output voltage generation: Analysis and
implementation,” in Proc. 33rd Annu. Conf. IEEE Ind. Electron. Soc.
(IECON), Nov. 2007, pp. 1524–1529.
[55] J. Reimers, L. Dorn-Gomba, C. Mak, and A. Emadi, “Automotive
traction inverters: Current status and future trends,” IEEE Trans. Veh.
Technol., vol. 68, no. 4, pp. 3337–3350, Apr. 2019.
[56] M. Duran, I. Gonzalez-Prieto, N. Rios-Garcia, and F. Barrero, “A sim-
ple, fast, and robust open-phase fault detection technique for six-phase
induction motor drives,” IEEE Trans. Power Electron., vol. 33, no. 1,
pp. 547–557, Jan. 2018.
[57] K. S. Khan, W. M. Arshad, and S. Kanerva, “On performance figures of
multiphase machines,” in Proc. 18th Int. Conf. Electr. Mach., Sep. 2008,
pp. 1–5.
[58] F. Baneira, J. Doval-Gandoy, A. G. Yepes, Ó. López, and
D. Pérez-Estévez, “Control strategy for multiphase drives with mini-
mum losses in the full torque operation range under single open-phase
fault,” IEEE Trans. Power Electron., vol. 32, no. 8, pp. 6275–6285,
Aug. 2017.
[59] F. Baneira, J. Doval-Gandoy, A. G. Yepes, Ó. López, and
D. Pérez-Estévez, “Comparison of postfault strategies for current ref-
erence generation for dual three-phase machines in terms of converter
losses,” IEEE Trans. Power Electron., vol. 32, no. 11, pp. 8243–8246,
Nov. 2017.
 1346 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019
[60] H. M. Eldeeb, A. S. Abdel-Khalik, and C. M. Hackl, “Postfault full
torque–speed exploitation of dual three-phase IPMSM drives,” IEEE
Trans. Ind. Electron., vol. 66, no. 9, pp. 6746–6756, Sep. 2019.
[61] A. Abdel-Khalik, M. I. Masoud, S. M. Gadoue, and B. W. Williams,
“Optimum flux distribution with harmonic injection for a multiphase
induction machine using genetic algorithms,” IEEE Trans. Energy
Convers., vol. 26, no. 2, pp. 501–512, Jun. 2011.
[62] A. S. Abdel-Khalik and S. M. Gadoue, “Improved flux pattern by third
harmonic injection for multiphase induction machines using neural
network,” Alexandria Eng. J., vol. 50, no. 2, pp. 163–169, Jun. 2011.
[63] Z. Nie, “Multiphase power electronic converters for electric vehicle
machine drive systems,” Ph.D. dissertation, Dept. Elect. Comput. Eng.,
McMaster Univ., Hamilton, ON, Canada, 2018.
[64] I. Zoric, M. Jones, and E. Levi, “Vector space decomposition algorithm
for asymmetrical multiphase machines,” in Proc. Int. Symp. Power
Electron. (Ee), Oct. 2017, pp. 1–6.
[65] Z. Wang, Y. Wang, J. Chen, and Y. Hu, “Decoupled vector space
decomposition based space vector modulation for dual three-phase
three-level motor drives,” IEEE Trans. Power Electron., vol. 33, no. 12,
pp. 10683–10697, Dec. 2018.
[66] H. Kouki, M. B. Fredj, and H. Rehaoulia, “Vector space decomposition
for double star induction machine modeling,” in Proc. 15th Int.
Conf. Sci. Techn. Autom. Control Comput. Eng. (STA), Dec. 2014,
pp. 581–586.
[67] V. Reber. (2016). E-Power: New Possibilities With 800-Volt Charg-
ing. Porsche Engineering Magazine. [Online]. Available: https://www.
porscheengineering.com/peg/en/about/magazine/
[68] A. Boglietti, R. Bojoi, A. Cavagnino, and A. Tenconi, “Efficiency
analysis of PWM inverter fed three-phase and dual three-phase induc-
tion machines,” in Proc. 41st IAS Annu. Meeting Conf. Rec. IEEE Ind.
Appl. Conf., vol. 1, Oct. 2006, pp. 434–440.
[69] A. Boglietti, R. Bojoi, A. Cavagnino, and A. Tenconi, “Efficiency
analysis of PWM inverter fed three-phase and dual three-phase high
frequency induction machines for low/medium power applications,”
IEEE Trans. Ind. Appl., vol. 55, no. 5, pp. 2015–2023, May 2008.
[70] A. R. Bakhshai, G. Joos, and H. Jin, “Space vector PWM control
of a split-phase induction machine using the vector classification
technique,” in Proc. 13th Annu. Appl. Power Electron. Conf. Expo.
(APEC), vol. 2, Feb. 1998, pp. 802–808.
[71] P. R. Rakesh and G. Narayanan, “Investigation on zero-sequence
signal injection for improved harmonic performance in split-phase
induction motor drives,” IEEE Trans. Ind. Electron., vol. 64, no. 4,
pp. 2732–2741, Apr. 2017.
[72] J. Prieto, E. Levi, F. Barrero, and S. Toral, “Output current ripple
analysis for asymmetrical six-phase drives using double zero-sequence
injection PWM,” in Proc. 37th Annu. Conf. IEEE Ind. Electron. Soc.
(IECON), Nov. 2011, pp. 3692–3697.
[73] P. R. Rakesh and G. Narayanan, “Analysis of sine-triangle and zero-
sequence injection modulation schemes for split-phase induction motor
drive,” IET Power Electron., vol. 9, no. 2, pp. 344–355, Oct. 2016.
[74] T. Zhao, Y. Zhang, H. Xu, X. Gui, and D. Xu, “DTP-PMSM drive
with proportional resonant regulator and dual zero injection PWM
control method,” in Proc. 21st Int. Conf. Electr. Mach. Syst. (ICEMS),
Oct. 2018, pp. 1628–1633.
[75] C. Wang, K. Wang, and X. You, “Research on synchronized SVPWM
strategies under low switching frequency for six-phase VSI-fed asym-
metrical dual stator induction machine,” IEEE Trans. Ind. Electron.,
vol. 63, no. 11, pp. 6767–6776, Nov. 2016.
[76] Y. Yu, L. Gao, Y. Liu, and F. Chai, “24-sector space vector decompo-
sition for a dual three-phase PMSM,” in Proc. 17th Int. Conf. Electr.
Mach. Syst. (ICEMS), Oct. 2014, pp. 1601–1606.
[77] K. Marouani, L. Baghli, D. Hadiouche, A. Kheloui, and A. Rezzoug,
“A new PWM strategy based on a 24-sector vector space decomposition
for a six-phase VSI-fed dual stator induction motor,” IEEE Trans. Ind.
Electron., vol. 55, no. 5, pp. 1910–1920, May 2008.
[78] K. Marouani, L. Baghli, D. Hadiouche, A. Kheloui, and A. Rezzoug,
“Discontinuous SVPWM techniques for double star induction motor
drive control,” in Proc. 32nd Annu. Conf. IEEE Ind. Electron. (IECON),
Nov. 2006, pp. 902–907.
[79] D. Yazdani, S. A. Khajehoddin, A. Bakhshai, and G. Joos, “A gener-
alized space vector classification technique for six-phase inverters,” in
Proc. IEEE Power Electron. Spec. Conf., Jun. 2007, pp. 2050–2054.
[80] D. Yazdani, S. A. Khajehoddin, A. Bakhshai, and G. Joos, “Full
utilization of the inverter in split-phase drives by means of a dual
three-phase space vector classification algorithm,” IEEE Trans. Ind.
Electron., vol. 56, no. 1, pp. 120–129, Jan. 2009.
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
[81] Y. He, Y. Wang, J. Wu, Y. Feng, and J. Liu, “A comparative study
of space vector PWM strategy for dual three-phase permanent-magnet
synchronous motor drives,” in Proc. 25th Annu. IEEE Appl. Power
Electron. Conf. Expo. (APEC), Feb. 2010, pp. 915–919.
[82] G. Grandi, G. Serra, and A. Tani, “Space vector modulation of a
six-phase VSI based on three-phase decomposition,” in Proc. Int.
Symp. Power Electron., Elect. Drives, Automat. Motion, Jun. 2008,
pp. 674–679.
[83] K. Koiwa and J.-I. Itoh, “A maximum power density design method
for nine switches matrix converter using SiC-MOSFET,” IEEE Trans.
Power Electron., vol. 31, no. 2, pp. 1189–1202, Feb. 2016.
[84] A. S. Salem, R. A. Hamdy, A. S. Abdel-Khalik, I. F. El-Arabawy, and
M. S. Hamad, “Performance of nine-switch inverter-fed asymmetrical
six-phase induction machine under machine and converter faults,” in
Proc. 18th Int. Middle East Power Syst. Conf. (MEPCON), Dec. 2016,
pp. 711–716.
[85] E. C. dos Santos, C. B. Jacobina, and O. I. da Silva, “Six-phase machine
drive system with nine-switch converter,” in Proc. 37th Annu. Conf.
IEEE Ind. Electron. Soc. (IECON), Nov. 2011, pp. 4204–4209.
[86] L. Pan et al., “A novel space-vector modulation method for nine-switch
converter,” IEEE Trans. Power Electron., to be published.
[87] N. Jarutus and Y. Kumsuwan, “Discontinuous and continuous space
vector modulations for a nine-switch inverter,” in Proc. IEEE 8th
Int. Power Electron. Motion Control Conf. (IPEMC-ECCE Asia),
May 2016, pp. 581–588.
[88] N. Jarutus and Y. Kumsuwan, “Novel modulation strategy based
on generalized two-level PWM theory for nine-switch inverter with
reduction of switching commutation,” in Proc. 42nd Annu. Conf. IEEE
Ind. Electron. Soc. (IECON), Oct. 2016, pp. 3529–3534.
[89] X. Li, L. Qu, B. Zhang, G. Zhang, and H. Liao, “A simplified
modulation strategy of nine-switch inverter to cut off half of switching
modes,” IEEE Access, vol. 6, pp. 7254–7261, 2018.
[90] N. Jarutus and Y. Kumsuwan, “A carrier-based phase-shift space vector
modulation strategy for a nine-switch inverter,” IEEE Trans. Power
Electron., vol. 32, no. 5, pp. 3425–3441, May 2017.
[91] N. Jarutus and Y. Kumsuwan, “A comparison between level- and phase-
shift space vector duty-cycle modulations using a nine-switch inverter
for an ASD,” in Proc. 18th Int. Conf. Elect. Mach. Syst. (ICEMS),
Oct. 2015, pp. 1877–1883.
[92] F. Gao, L. Zhang, D. Li, P. C. Loh, Y. Tang, and H. Gao, “Optimal
pulsewidth modulation of nine-switch converter,” IEEE Trans. Power
Electron., vol. 25, no. 9, pp. 2331–2343, Sep. 2010.
[93] S. M. Dehghan, M. Mohamadian, and M. A. E. Andersen, “Full
space vectors modulation for nine-switch converters including CF &
DF modes,” in Proc. 4th Int. Power Eng. Optim. Conf. (PEOCO),
Jun. 2010, pp. 89–94.
[94] P. R. U. Guazzelli et al., “Dual predictive current control of grid
connected nine-switch converter applied to induction generator,” in
Proc. 13th IEEE Int. Conf. Ind. Appl. (INDUSCON), Nov. 2018,
pp. 1038–1044.
[95] S. Sharma, M. Aware, and A. Bhowate, “Direct torque control of
symmetrical six-phase induction machine using nine switch inverter,” in
Proc. IEEE Transp. Electrific. Conf. (ITEC-India), Dec. 2017, pp. 1–6.
[96] O. Gulbudak and M. Gokdag, “Predictive dual-induction machine
control using nine-switch inverter for multi-drive systems,” in Proc.
IEEE 12th Int. Conf. Compat., Power Electron. Power Eng. (CPE-
POWERENG), Apr. 2018, pp. 1–6.
[97] S. M. Dehghan, M. Mohamadian, A. H. Rajaei, and A. Yazdian, “Full
modeling a series HEV based on BZS-NSI,” in Proc. 2nd Power
Electron., Drive Syst. Technol. Conf., Feb. 2011, pp. 439–444.
[98] S. M. Dehghan, M. Mohamadian, and A. Yazdian, “Hybrid electric
vehicle based on bidirectional Z-source nine-switch inverter,” IEEE
Trans. Veh. Technol., vol. 59, no. 6, pp. 2641–2653, Jun. 2010.
[99] K. Oka, Y. Nozawa, R. Omata, K. Suzuki, A. Furuya, and
K. Matsuse, “Characteristic comparison between five-leg inverter
and nine-switch inverter,” in Proc. Power Convers. Conf.-Nagoya,
Apr. 2007, pp. 279–283.
[100] C. B. Jacobina, I. S. de Freitas, C. R. da Silva, M. B. de Rossiter
Correa, and E. R. C. da Silva, “Reduced switch-count six-phase AC
motor drive systems without input reactor,” IEEE Trans. Ind. Electron.,
vol. 55, no. 5, pp. 2024–2032, May 2008.
[101] N. A. Azeez, J. Mathew, K. Gopakumar, and C. Cecati, “A 5th
and 7th order harmonic suppression scheme for open-end winding
asymmetrical six-phase IM drive using capacitor-fed inverter,” in Proc.
39th Annu. Conf. IEEE Ind. Electron. Soc. (IECON), Nov. 2013,
pp. 5118–5123.
 SALEM AND NARIMANI: REVIEW ON MPDs FOR AUTOMOTIVE TRACTION APPLICATIONS
[102] J. Loncarski, M. Leijon, C. Rossi, M. Srndovic, and G. Grandi,
“Current ripple evaluation in dual three-phase inverters for open-end
winding EV drives,” in Proc. Int. Conf. Connected Vehicles Expo
(ICCVE), Nov. 2014, pp. 507–513.
[103] M. Jones, F. Patkar, and E. Levi, “Carrier-based pulse-width modulation
techniques for asymmetrical six-phase open-end winding drives,” IET
Electr. Power Appl., vol. 7, no. 6, pp. 441–452, Jul. 2013.
[104] F. Patkar, A. Jidin, E. Levi, and M. Jones, “Performance comparison
of symmetrical and asymmetrical six-phase open-end winding drives
with carrier-based PWM,” in Proc. 6th Int. Conf. Elect. Eng. Inform.
(ICEEI), Nov. 2017, pp. 1–6.
[105] Z. Wang, J. Chen, and M. Cheng, “Modeling and control of neutral-
point-clamping (NPC) three-level inverters fed dual-three phase PMSM
drives,” in Proc. IEEE Energy Convers. Congr. Expo. (ECCE),
Sep. 2015, pp. 6565–6572.
[106] Z. Wang, J. Chen, M. Cheng, and R. Na, “Vector space decomposition
based control of neutral-point-clamping (NPC) three-level inverters fed
dual three-phase PMSM drives,” in Proc. 42nd Annu. Conf. IEEE Ind.
Electron. Soc. (IECON), Oct. 2016, pp. 2988–2993.
[107] S. Bhattacharya, D. Mascarella, G. Joós, J. M. Cyr, and J. Xu,
“A dual three-level T-NPC inverter for high-power traction applica-
tions,” IEEE J. Emerg. Sel. Topics Power Electron., vol. 4, no. 2,
pp. 668–678, Jun. 2016.
[108] Z. Wang, X. Wang, M. Cheng, and W. Hua, “Direct torque control of
T-NPC three-level inverter fed dual three-phase PMSM drives based
on double SVM strategy,” in Proc. 19th Int. Conf. Elect. Mach. Syst.
(ICEMS), Nov. 2016, pp. 1–7.
[109] J. Wen and K. M. Smedley, “Hexagram inverter for medium-voltage
six-phase variable-speed drives,” IEEE Trans. Ind. Electron., vol. 55,
no. 6, pp. 2473–2481, Jun. 2008.
[110] V. Oleschuk, R. Gregor, J. Rodas, and A. Giangreco, “Dual three-
phase system based on four neutral-point-clamped inverters with syn-
chronized space-vector modulation,” in Proc. 15th Eur. Conf. Power
Electron. Appl. (EPE), Sep. 2013, pp. 1–10.
[111] V. F. M. B. Melo, C. B. Jacobina, and N. Rocha, “Hybrid open-end
and NPC AC six-phase machine drive systems,” in Proc. IEEE Energy
Convers. Congr. Expo. (ECCE), Sep. 2014, pp. 3872–3879.
[112] R. V. Nair, K. Gopakumar, and L. G. Franquelo, “A very high resolution
stacked multilevel inverter topology for adjustable speed drives,” IEEE
Trans. Ind. Electron., vol. 65, no. 3, pp. 2049–2056, Mar. 2018.
[113] R. V. Nair, R. Chattopadhyay, S. Parashar, S. Bhattacharya, and
K. Gopakumar, “Cascaded active neutral point clamped and flying
capacitor inverter topology for induction motor drives applications,”
in Proc. IEEE Energy Convers. Congr. Expo. (ECCE), Sep. 2018,
pp. 6696–6702.
[114] R. V. Nair, Arun Rahul S, S. Pramanick, K. Gopakumar, and
L. G. Franquelo, “Novel symmetric six-phase induction motor drive
using stacked multilevel inverters with a single DC link and neutral
point voltage balancing,” IEEE Trans. Ind. Electron., vol. 64, no. 4,
pp. 2663–2670, Apr. 2017.
[115] M. Jones, S. N. Vukosavic, D. Dujic, and E. Levi, “A synchronous
current control scheme for multiphase induction motor drives,” IEEE
Trans. Energy Convers., vol. 24, no. 4, pp. 860–868, Dec. 2009.
[116] R. Bojoi, F. Farina, A. Tenconi, F. Profumi, and E. Levi, “Dual three-
phase induction motor drive with digital current control in the stationary
reference frame,” Power Eng., vol. 20, no. 3, pp. 40–43, Jun. 2006.
[117] A. G. Yepes, J. Malvar, A. Vidal, O. López, and J. Doval-Gandoy,
“Current harmonics compensation based on multiresonant control in
synchronous frames for symmetrical n-phase machines,” IEEE Trans.
Ind. Electron., vol. 62, no. 5, pp. 2708–2720, May 2015.
[118] J. Karttunen, S. Kallio, P. Peltoniemi, and P. Silventoinen, “Current har-
monic compensation in dual three-phase PMSMs using a disturbance
observer,” IEEE Trans. Ind. Electron., vol. 63, no. 1, pp. 583–594,
Jan. 2016.
[119] Y. Hu, Z. Q. Zhu, and K. Liu, “Current control for dual three-
phase permanent magnet synchronous motors accounting for current
unbalance and harmonics,” IEEE J. Emerg. Sel. Topics Power Electron.,
vol. 2, no. 2, pp. 272–284, Jun. 2014.
[120] A. Negahdari, A. G. Yepes, J. Doval-Gandoy, and H. A. Toliyat,
“Efficiency enhancement of multiphase electric drives at light-load
operation considering both converter and stator copper losses,” IEEE
Trans. Power Electron., vol. 34, no. 2, pp. 1518–1525, Feb. 2019.
[121] TM4. Accessed: Nov. 20, 2019. [Online]. Available: https://www.
tm4.com/products/direct-drive-electric-powertrain/sumo-md/
Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.
1347
[122] D. Semenov, B. Tian, Q.-T. An, and L. Sun, “Position estimation for
sensorless FOC of five-phase PMSM in electric vehicles,” in Proc.
Australas. Univ. Power Eng. Conf. (AUPEC), Sep. 2016, pp. 1–5.
[123] S. Rhaili, A. Abbou, A. Ziouh, and R. Elidrissi, “Comparative study
between PI and FUZZY logic controller in vector controlled five-
phase PMSG based variable-speed wind turbine,” in Proc. IEEE 12th
Int. Conf. Compat., Power Electron. Power Eng. (CPE-POWERENG),
Apr. 2018, pp. 1–6.
[124] S. Kouro, P. Cortes, R. Vargas, U. Ammann, and J. Rodriguez, “Model
predictive control—A simple and powerful method to control power
converters,” IEEE Trans. Ind. Electron., vol. 56, no. 6, pp. 1826–1838,
Jun. 2009.
[125] F. Barrero, M. R. Arahal, R. Gregor, S. Toral, and M. J. Duran,
“A proof of concept study of predictive current control for VSI-
driven asymmetrical dual three-phase AC machines,” IEEE Trans. Ind.
Electron., vol. 56, no. 6, pp. 1937–1954, Jun. 2009.
[126] H. Lu, J. Li, R. Qu, D. Ye, and Y. Lu, “Fault-tolerant predictive
control of six-phase PMSM drives based on pulsewidth modula-
tion,” IEEE Trans. Ind. Electron., vol. 66, no. 7, pp. 4992–5003,
Jul. 2019.
[127] D. Ye, J. Li, R. Qu, H. Lu, and Y. Lu, “Finite set model predictive
MTPA control with VSD method for asymmetric six-phase PMSM,”
in Proc. IEEE Int. Electr. Mach. Drives Conf. (IEMDC), May 2017,
pp. 1–7.
[128] H. Lu, J. Li, R. Qu, D. Ye, Y. Lu, and R. Zhang, “Post-fault
model predictive control of asymmetrical six-phase permanent magnet
machine with improved mathematical model,” in Proc. IEEE Int. Electr.
Mach. Drives Conf. (IEMDC), May 2017, pp. 1–7.
[129] I. Gonzalez-Prieto, M. J. Duran, J. J. Aciego, C. Martin, and F. Barrero,
“Model predictive control of six-phase induction motor drives using
virtual voltage vectors,” IEEE Trans. Ind. Electron., vol. 65, no. 1,
pp. 27–37, Jan. 2018.
[130] Y. Luo and C. Liu, “A simplified model predictive control for a dual
three-phase PMSM with reduced harmonic currents,” IEEE Trans. Ind.
Electron., vol. 65, no. 11, pp. 9079–9089, Nov. 2018.
[131] E. Jung, H. Yoo, S.-K. Sul, H.-S. Choi, and Y.-Y. Choi, “A nine-
phase permanent-magnet motor drive system for an ultrahigh-speed
elevator,” IEEE Trans. Ind. Appl., vol. 48, no. 3, pp. 987–995,
May/Jun. 2012.
[132] P. Brockerhoff, W. Schön, P. Blaha, P. Václavek, and Y. Burkhardt,
“Disc inverter in highly integrated 9-phase drivetrain for E-mobility,”
in Proc. 17th Eur. Conf. Power Electron. Appl. (EPE ECCE-Eur.),
Sep. 2015, pp. 1–9.
[133] L. Verkroost, J. Van Damme, H. Vansompel, F. De Belie, and
P. Sergeant, “Module connection topologies and interleaving strategies
for integrated modular motor drives,” in Proc. IEEE Int. Electr. Mach.
Drives Conf. (IEMDC), May 2019, pp. 559–564.
[134] J. Van Damme, L. Verkroost, H. Vansompel, F. De Belie, and
P. Sergeant, “A holistic DC link architecture design method for mul-
tiphase integrated modular motor drives,” in Proc. IEEE Int. Electr.
Mach. Drives Conf. (IEMDC), May 2019, pp. 1593–1598.
[135] I. Subotic, E. Levi, M. Jones, and D. Graovac, “An integrated battery
charger for EVs based on an asymmetrical six-phase machine,” in
Proc. 39th Annu. Conf. IEEE Ind. Electron. Soc. (IECON), Nov. 2013,
pp. 7244–7249.
[136] I. Subotic, E. Levi, and N. Bodo, “A fast on-board integrated battery
charger for EVs using an asymmetrical six-phase machine,” in Proc.
IEEE Vehicle Power Propuls. Conf. (VPPC), Oct. 2014, pp. 1–6.
[137] I. Subotic, N. Bodo, and E. Levi, “Single-phase on-board integrated
battery chargers for EVs based on multiphase machines,” IEEE Trans.
Power Electron., vol. 31, no. 9, pp. 6511–6523, Sep. 2016.
[138] U.S. DRIVE Electrical and Electronics Technical Team Roadmap,
Office Energy Efficiency Renew. Energy, Washington, DC, USA,
Oct. 2017.
[139] A. Bindra, “Wide-bandgap-based power devices: Reshaping the power
electronics landscape,” IEEE Power Electron. Mag., vol. 2, no. 1,
pp. 42–47, Mar. 2015.
[140] S. Hazra, S. Madhusoodhanan, G. Karimi Moghaddam, K. Hatua,
and S. Bhattacharya, “Design considerations and performance eval-
uation of 1200-V 100-A SiC MOSFET-based two-level voltage source
converter,” IEEE Trans. Ind. Appl., vol. 52, no. 5, pp. 4257–4268,
Sep./Oct. 2016.
[141] J. Liu, W. Su, X. Tai, W. Sun, L. Gu, and X. Wen, “Development of
an inverter using hybrid SIC power module for EV/HEV applications,”
in Proc. 19th Int. Conf. Elect. Mach. Syst. (ICEMS), Nov. 2016,
pp. 1–5.
 1348 IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION, VOL. 5, NO. 4, DECEMBER 2019

[142] H. Zheng, X. Wang, X. Wang, L. Ran, and B. Zhang, “Using SiC Mehdi Narimani (S’09–M’13–SM’15) received the
MOSFETs to improve reliability of EV inverters,” in Proc. IEEE 3rd Ph.D. degree in electrical engineering from the Uni-
Workshop Wide Bandgap Power Devices Appl. (WiPDA), Nov. 2015, versity of Western Ontario, London, ON, Canada,
pp. 359–364. in 2012.
[143] H. Wang and D. Jiang, “Design of high temperature gate driver for He was a Power Electronics Engineer with Rock-
SiC MOSFET for EV motor drives,” in Proc. IEEE Transp. Electrific. well Automation Canada, Cambridge, ON. He is
Conf. Expo, Asia–Pacific (ITEC Asia–Pacific), Aug. 2017, pp. 1–6. currently an Assistant Professor with the Department
[144] T. Shimomura, K. Numakura, D. Sato, and T. Hayashi, “High speed of Electrical and Computer Engineering, McMas-
dV/dt control technology for a SiC power module for EV/HEV invert- ter University, Hamilton, ON. He has authored or
ers using a multistage drive circuit,” in Proc. IEEE Energy Convers. coauthored more than 90 journals and conference
Congr. Expo. (ECCE), Sep. 2018, pp. 3542–3546. proceeding papers, coauthored a Wiley-IEEE press
book, and holds seven issued/pending U.S./European patents. His current
research interests include power conversion, high-power converters, control
Ahmed Salem (S’19) received the B.Sc. and M.Sc.
of power electronics converters, fast EV chargers, and wireless EV charging
degrees in electrical engineering from Alexandria systems.
University, Alexandria, Egypt, in 2014 and 2018,
respectively. He is currently pursuing the Ph.D.
degree with the Department of Electric and
Computer Engineering, McMaster University,
Hamilton, ON, Canada.
He is a member of the team of the Center for
Mechatronics and Hybrid Technology (CMHT),
McMaster University. His work is on the E-Motor
part of the EECOMOBILITY CREATE/ORF
project. His research interests include power electronics and electric drives
in electrified transportation applications.

Authorized licensed use limited to: b-on: UNIVERSIDADE DA BEIRA INTERIOR. Downloaded on September 17,2020 at 10:39:28 UTC from IEEE Xplore. Restrictions apply.