IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO.

4, JULY/AUGUST 2021 4143

Analysis, Design, and Comparison of V2V Chargers

for Flexible Grid Integration
Emin Ucer , Student Member, IEEE, Ramona Buckreus , Student Member, IEEE,
Md Ehsanul Haque , Student Member, IEEE, Mithat Kisacikoglu , Member, IEEE,
Yilmaz Sozer , Senior Member, IEEE, Salman Harasis , Student Member, IEEE,
Mustafa Guven, Senior Member, IEEE, and Luigi Giubbolini

Abstract—This study proposes a flexible, compact, efficient, and
cost effective electric vehicle-to-electric vehicle charge sharing so-
lution that will lead to faster and wider customer adoption of
electric vehicles (EVs) as an alternative to current grid-to-vehicle
charging methodologies. The energy will be transferred between
EVs using a bidirectional dc–dc converter in a conductive way,
which can take place at parking lots of workplaces, campuses, or
residential premises and highways. The proposed design provides
compact infrastructure, wide input and output voltage ranges with
bidirectional buck–boost operation, and fast power transfer com-
pared to Level 2 charging stations. A complete design comparison
between Si and SiC based converters has been carried out based
on the available power modules and magnetic cores. The compar-
ison demonstrates the effect of switching frequency and switch
selection on the size of passive components, the converter efficiency,
and the power density of the converter. The developed study eval-
uated three different bidirectional dc–dc converter topologies and
validated the developed prototype with experimental results.
Index Terms—Electric vehicles (EVs), EV chargers, SiC charger
design, electric vehicle-to-electric vehicle (V2V) charging.
I. INTRODUCTION
ODAY, real-time charging from the utility grid is rec-
T ognized to be the mainstream way of “fueling” electric
vehicles (EVs). However, since the current EV penetration rates
Manuscript received April 8, 2020; revised September 3, 2020, December 20,
2020, and March 25, 2021; accepted May 18, 2021. Date of publication May 27,
2021; date of current version July 16, 2021. Paper 2019-TSC-1372.R3, presented
at the 2019 IEEE Transportation Electrification Conference and Expo, Detroit,
MI, USA, Jun. 19–21, and approved for publication in the IEEE TRANSACTIONS
ON INDUSTRY APPLICATIONS by the Transportation Systems Committee of the
IEEE Industry Applications Society. This work was supported in part by the
National Science Foundation under Award 1755996 and in part by the Center for
Advanced Vehicle Technologies, The University of Alabama. (Corresponding
author: Mithat Kisacikoglu.)
Emin Ucer, Ramona Buckreus, and Mithat Kisacikoglu are with the
Department of Electrical and Computer Engineering, The University of
Alabama, Tuscaloosa, AL 35487 USA (e-mail: eucer@crimson.ua.edu; rbuck-
reus@crimson.ua.edu; mkisacik@ua.edu).
Md Ehsanul Haque, Yilmaz Sozer, and Salman Harasis are with the De-
partment of Electrical and Computer Engineering, The University of Akron,
Akron, OH 32816 USA (e-mail: mh248@zips.uakron.edu; ys@uakron.edu;
skh60@uakron.edu).
Mustafa Guven is with the Caterpillar, Inc., Mossville, IL 60015 USA (e-mail:
guven_mustafa_k@cat.com).
Luigi Giubbolini is with the Andromeda Power, LLC, Long Beach, CA 90815
USA (e-mail: luigi@andromedapower.com).
Color versions of one or more figures in this article are available at https:
//doi.org/10.1109/TIA.2021.3084576.
Digital Object Identifier 10.1109/TIA.2021.3084576
are very limited, many of the problems—such as increased dis-
tribution level peak demand tied to grid integration of EVs—are
not clearly differentiated [2]–[4]. Studies show that residential
Level 2 EV charging will result in disruptive problems in the
future distribution grid if not managed well [2]–[9]. This phe-
nomenon requires tightly controlled residential charging man-
agement solutions not to violate distribution system constraints,
which limit EV charging power and increase charging time.
Therefore, residential EV charging must be accompanied by fast
dc charging stations that can be connected at the medium voltage
level and have less impact on the distribution grid operation
to sustain the EV growth [10]. Approximately 40 million EVs
expected to be on the road in the US by 2030 will require 400-K
new dc fast charging (DCFC) outlets [11]. Furthermore, an
average US driver operates his/her vehicle 59.69 min/d [12] and
requests DCFC mainly around noon and between 3–6 P.M. [13].
Consequently, to respond to this tight spatio-temporal charging
request within a city, an upfront cost over $90 K per DCFC
station is needed [14]. Furthermore, issues around land owner-
ship, municipal permitting, electrical grid limitations, and other
factors have slowed the DCFC installations [15].
Electric vehicle-to-electric vehicle (V2V) charging technol-
ogy (a.k.a. peer-to-peer charging) is proposed to offer solutions
to the increased energy demand. It is considered as a new way
of transferring energy between vehicles without directly relying
on the utility grid. Some car manufacturers, such as Rivian
and Hyundai, announced to include this feature in their new
EV models [16]–[18]. In this study, we propose a V2V charge
sharing network (CSN) philosophy to provide an alternative,
more convenient, and flexible way of conducting EV charging.
The design and implementation of V2V CSN will greatly reduce
the range anxiety of EV owners with minimal infrastructure cost.
The proposed solution has the potential to provide an invaluable
service that will benefit EV owners, local communities and
municipalities, and the utility grid. Another major feature of
the proposed V2V scheme is that V2V is an indirect demand
response management tool that works mainly during peak times.
Such a scenario is very viable since most EV owners charge
their EVs daily at home after only using it for 25–30 mi of
commute. Therefore, they have, on average, about 50%–60% of
their battery capacity available to sell [19]. V2V CSN will allow
users (either battery EV, plug-in hybrid, or fuel-cell EV) with
unused electric energy on their board to connect with users who

0093-9994 © 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See https://www.ieee.org/publications/rights/index.html for more information.

Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4144
need to charge their vehicles at comparable transfer rates to fast
charging stations.
V2V charging requires an analysis to figure out how to
pair suppliers to receivers with efficient matching algorithms
and how to enable energy exchange with current EV charging
technologies. Authors previously developed a Java-based sim-
ulation tool to analyze the viability of the proposed V2V CSN
technology [1], [20], [21]. The tool generates a customizable
simulation environment with different parameters, including
EV types and counts, charging station types and locations, and
user mobility patterns. Some recent papers also addressed the
challenges surrounding energy sharing between EVs. In [22],
Zhang et al. present a matching algorithm to facilitate V2V coop-
erative energy transfer. It compares two different V2V matching
algorithms with traditional grid-based charging. In [23], the
matching of demander EVs to both V2V suppliers and existing
charging stations has been studied in an efficient and privacy
preserving manner. In [24], an authentication protocol between
vehicles before they actually start V2V charging has been pre-
sented.
On the other hand, the concept of dc–dc power conversion
for automotive applications has been addressed in the liter-
ature for various applications [25], [26]. Recently, three dif-
ferent solutions for V2V energy transfer have been compared
in [27]: vehicle-to-grid and grid-to-vehicle (V2G+G2V), V2V
over direct ac interconnection, and V2V over dc interconnection
(dcV2V). It is shown that dcV2V is more efficient than the
other options due to the reduced number of energy conversions.
Moreover, a combined half-bridge (CHB) type converter is
used for dcV2V, which has previously been reported in [28].
In [28], mobile charging with an energy storage device using
a three-phase, interleaved, bidirectional, cascaded buck–boost
dc–dc converter was reported and compared with CHB design
counterpart. However, the study does not completely outline
technical challenges and design procedures specific to V2V
energy transfer. A converter design is also proposed in [29],
which allows four common energy transfer modes, namely V2G,
G2V, vehicle-to-home, and V2V. The authors focus on higher
power density design with efficient operation for the afore-
mentioned operation modes. V2V operation in [29] proposes
to use on-board chargers, which limits the speed of energy
transfer.
The efficiency of the power conversion is another critical
factor to consider for the adaptation and wide-spread usage of
V2V charging. This vital requisite can be accomplished by using
wide band-gap (WBG) devices, which have faster switching
and lower loss capability [28], [30]. The availability of WBG
devices enables the fulfillment of highly efficient, reliable, and
compact power electronic systems [31]–[33]. High switching
frequency operation of WBG devices helps to reduce the size
of passive components, which, in turn, increases the volumetric
power density of the charging stations. Designing a high volu-
metric power density and lighter power converter are significant
features for the V2V operation. Furthermore, the overall loss of
WBG devices, such as silicon carbide (SiC) MOSFET, is lower,
the efficiency is higher, and the required heat sink size is smaller.
These features reduce the thermal requirement and the weight
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021
of the total system to increase the volumetric and gravimetric
power densities of the charger.
For the comparison of conventional Si devices and WBG SiC
devices, the available literature is explored. In [34] and [35], the
performance of Si and SiC MOSFET based isolated interleaved
boost dc–dc converter is compared for a microinverter applica-
tion. Ho et al. in [36] have shown the performance of interleaved
boost converter using Si and SiC diodes for PV application,
where higher efficiency is achieved due to absence of reverse
recovery current in SiC schottky diodes. In [37], Öztürk has
shown different design steps for three-phase interleaved boost
converter using SiC devices for an EV application. In [38] and
[39], multiphase bidirectional dc–dc converters are proposed for
two energy storage systems of different levels, and several ad-
vantages are discussed for the interleaving feature. Neutral point
clamped three-level interleaved nonisolated dc–dc converter is
proposed for EV charging system as another converter topology
with reduced stress on the devices [40].
The unique challenge of this study differs from the literature
in that the supplier and receiver voltages vary greatly depending
on the battery state-of-charge (SOC) levels and can even overlap
each other during the energy sharing. Furthermore, the V2V
charger needs to be directionally agnostic providing utmost
flexibility for the customer usage so that any side of the V2V
charger can be supplier/receiver.
Our main contributions in this article can be summarized as
follows.
1) Performing a complete analysis for the considered topolo-
gies in terms of total effective inductor current ripple based
on passive components at different switching frequencies.
2) Designing bidirectional interleaved single-phase, two-
phase, and three-phase converters (buck–boost) for flexi-
ble V2V charging and discussing their application specific
operation modes.
3) Presenting the closed-loop current controller design and
their respective simulation results.
4) Verifying the operation by performing hardware experi-
ments for single- and two-phase boost–buck interleaved
converters.
Section II of this article presents an analysis of V2V energy
transfer for different dc–dc converter topologies. Section III
presents the converter design and comparison study for different
options. Section IV presents simulation results to verify the
analysis performed in Section III. Section V demonstrates the
controller design for the experimental setup. Section VI presents
the experimental verification for Si insulated-gate bipolar tran-
sistor (IGBT) design. Finally, Section VII concludes this article.
II. ANALYSIS OF V2V ENERGY TRANSFER USING
DC–DC CONVERTERS
Three bidirectional (interleaved) dc–dc converter topologies
shown in Fig. 1 are reported in the literature as candidates for
similar applications [28], [41], [42]. Each V2V charger solution
presented in Fig. 1 consists of a power stage with an appropriate
set of gate-drive circuits, a digital signal processor (DSP), volt-
age sensing circuits to measure the local EV battery voltages
UCER et al.: ANALYSIS, DESIGN, AND COMPARISON OF V2V CHARGERS FOR FLEXIBLE GRID INTEGRATION 4145

Fig. 1. Bidirectional dc–dc converters for V2V charger. (a) Single phase. (b) Two phase. (c) Three phase. (d) Hardware setup.

TABLE I TABLE II
OPERATION MODES OF SINGLE-PHASE BIDIRECTIONAL CONVERTER DUTY CYCLE, INDUCTOR CURRENT RIPPLE, AND INDUCTANCE FOR
CONTINUOUS BOOST AND BUCK OPERATION MODES

(V1 and V2 ), and one current sensor to measure the real-time

phase current (ILa ). The DSP communicates with the vehicles
via CAN through a data link embedded in the charging cable
to coordinate the two ends. The direction of charge transfer is
determined by the on-board DSP based on the users’ commands.
The charger has four separate operational modes, identified
TABLE III
in Table I for the sample case of single-phase conversion [see TOTAL RIPPLE CURRENT IN CONT. BOOST AND BUCK OPERATIONS [45], [46]
Fig. 1(a)]. Since the buck and boost mode cannot occur at the
same time, only one phase-leg operates at switching mode while
the other is in static mode. For instance, referring to Fig. 1(a),
there is one switch that is actively modulating the flow of power
between the vehicles (active) and one switch that is operating
as a synchronous rectifier (SR) for the case of a MOSFET. The
remaining two switches are fixed in a constant state (ON/OFF) to
statically attach the sink (buck mode) or source (boost mode).
In multiple-phase operation, as shown in Fig. 1(b) and (c),
the switching signals for the parallel, active switches are shifted
by T /N , where T is the switching period, and N is the number
of parallel phases. This corresponds to a phase shift of 180◦
and 120◦ for two- and three-phase operation, respectively. This
operation is called interleaving. Interleaving the dc–dc converter seen at an inductor (La,b,c in Fig. 1) ΔIL , and the required
brings in advantages in terms of reducing passive filter require- inductance value for different design metrics for boost and buck
ments at the expense of more circuit complexity [43], [44]. operation. As seen in Table II, the current ripple at one inductor
Another advantage is that the peak phase current reduces with not only depends on the operation mode but also on the switching
interleaving by 1/N . Therefore, maximum average power that frequency and inductance value. The inductance value can be
can be transferred increases for a given semiconductor device determined to meet a certain ripple requirement at a particular
rating. output current, i.e., ΔIL ÷ Iout . However, this only holds true for
The reduction of the peak-to-peak
 current ripple can be seen single-phase
 topology as for multiphase topologies, the effective
at the input (ΔIin = Δ  IL in Fig. 1) during boost mode or ripple Δ IL is not equal to the ripple measured at an inductor
 (ΔIout = Δ IL in Fig. 1) during buck mode,
at the output ΔIL . Thus, the inductance value is determined to meet an
where Δ IL is the ripple of the sum of all inductor currents effective ripple requirement of Δ IL ÷ Iout .
and, thus, the effective ripple of the converter. Table II provides Table III further demonstrates how the effective current ripple

the equations for the duty cycle, the resulting current ripple Δ IL reduces depending on the number of parallel phases

Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4146
Fig. 2. Required inductance relative  to input and output voltages for
Pout =27 kW, fs =20 kHz, and Δ IL ÷ Iout =10% with (a) one-phase,
(b) two-phase, and (c) three-phase configuration.
N (and thus the phase shift angle) and the duty
 cycle d. This
relationship between the effective ripple Δ IL and the rip-
ple at an inductor ΔIL is denoted as f (N, d). The function
f (N, d) is equal to unity for single-phase topology. It can be
seen that the required inductance for a given ripple specification
at input (boost mode) and output (buck mode) is reduced by
using interleaved multiphase converter topology. However, the
current ripple and required inductance highly depend on the
particular operation point. As battery voltages of today’s EVs
can vary greatly, and they show an increasing trend, a case
analysis is done for 250 V V < Vin < 800 V and 250 V < Vout <
800 V to allow flexible usage of the V2V charger with a wide
range of EV battery voltages. The required inductance for an
output power of 27 kW, a switching frequency of 20 kHz, and a
peak-to-peak ripple of 10% is shown in Fig. 2 for three different
converter cases. This figure shows that the required inductance
does not strictly decrease with the number of phases but depends
on the particular operation point. The highest reduction in the
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021
TABLE IV
 FOR DIFFERENT CONVERTER CONFIGURATIONS WITH
INDUCTANCE NEEDED
Δ IL ÷ IOUT = 10% AND POUT =27 KW
inductance occurs at d = 0.5 for two-phase operation and d =
0.33 or d = 0.66 for three-phase operation. At these points,
the ripples of the inductor currents cancel each other. On the
other hand, EV battery voltages during the energy exchange
can be very dependent on the EV model and SOC of the
supplier/receiver battery making it very hard to design the con-
verter at a fixed duty ratio. Therefore, the minimum required
inductance that covers all the operating regions in Fig. 2 while
always satisfying 10% ripple requirement is the most stringent
requirement on the inductance design. That is termed as Lmax
in Fig. 2 and is equal to 2956 μ H for single-phase, 1478 μH
for two-phase, and 985-μH three-phase configurations for this
particular case study. A similar study can also be done on higher
switching frequencies but is not presented here.
III. DESIGN COMPARISON OF 27-KW FAST V2V CHARGER
The design and implementation of V2V charger should reduce
the range anxiety of EVs with minimal infrastructure cost and
footprint. The users should easily store the unit in the trunk
of their EVs and move the unit around easily. Therefore, we
selected a 27-kW output power rating for our design, which is
also close to the power rating of a commercial DCFC express
charging device [47]. This rating can approximately give 30 mi
of EV range in 20 min1 with a compact footprint as will be
investigated in this section. Furthermore, designs of single-,
dual-, and three-phase interleaved dc–dc converters are devel-
oped and compared using Si and SiC semiconductor options in
this section.
A. Inductor Design
The modeling considers the operating voltages that require
the highest inductance values illustrated in Fig. 2 for the three
different converters. The phase inductor values are calculated for
different switching frequencies considering 10% peak-to-peak
ripple current. Accordingly, the largest inductor values required
to meet the design specifications are listed in Table IV. The
current ripples will be less than 10% at all other voltages in
the analyzed operating regions. The input and output voltage
ripples are assumed to be small and negligible. As can be seen
from Table IV, increasing the switching frequency from 20 to
100 kHz reduces the inductance requirements to 20% of the orig-
inal value (i.e., L100 kHz /L20 kHz = 20%) as expected. Therefore,
high switching frequency operation offers a competitive solution
1 The energy efficiency of the EV is assumed to be 30 kWh/100 mi. This is
equal to 112 MPGe assuming that 33.7 kWh of energy is equivalent to one gallon
of gas.
UCER et al.: ANALYSIS, DESIGN, AND COMPARISON OF V2V CHARGERS FOR FLEXIBLE GRID INTEGRATION
TABLE V
DESIGN SUMMARY OF DIFFERENT INDUCTOR MODELS
Fig. 3. Core sizing at different frequencies.
to reduce the size and the weight of the passive components,
which can be enabled by using SiC switches. However, this
requires a closer look into the problem, which is presented as
follows.
For an accurate size comparison of the configurations oper-
ating at various switching frequencies, different core models
with different inductance values are considered for the design
realization. The inductor design is carried out based on the
inductor geometry and the material property, and are given
considering the following equations [31], [48]:
0.4πμ · N 2 · A · 10−2
L= (1)
l other hand, multiphase topologies allow higher inductor current
0.4π · N · Imax
Hmax = (2)
l ing trend of core loss is observed with higher number of phases
Erms · 108
Bmax = (3)
4.44 · f · A · N
where N , μ, A, l, f , Imax , and Erms are the number of turns, the
permeability of the core material, the cross-sectional area of the
core, the core length, operating frequency, the maximum phase
current, and the rms value of the induced voltage, respectively.
Table V presents the inductor design summary using different
core models. Powder iron cores are considered for their higher dc
bias capabilities and low losses. The maximum current through
the inductor depends on the converter configurations. As the
permeability of the core is highly dependent on the dc bias,
single-phase configuration requires a larger size core compared
to multiphase topologies to achieve nominal inductance value
at their rated current. Furthermore, the value of inductance
decreases with higher number of phases and higher switching
frequencies. Fig. 3 depicts the effective volume comparison of
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4147
Fig. 4. Total inductor loss calculation at different switching frequencies.
the selected powder iron cores at different switching frequen-
cies for the total number of cores in each design. The figure
shows that moving from single- to three-phase configuration
via implementing the interleaving, the total core volume can
be significantly reduced. Moreover, increasing the switching
frequency decreases the resulting volume further. Therefore, we
see that high switching operation using a multiphase SiC-based
converter helps reduce the inductor size significantly, providing
a compact and flexible V2V operation.
On the other hand, to compare the effect of the switching
frequency on the inductor power loss, copper and core losses
have been calculated separately. Fig. 4 compares the total losses
of the inductors at different switching frequencies based on the
selected core for each configuration. Having higher number of
phases and/or higher switching frequency reduce the inductance
requirement, which gives the opportunity to use smaller size core
along with a smaller number of turns. Both criteria help reducing
the total wire length and, hence, the dc resistance of the coil.
Therefore, the inductor copper loss has a decreasing trend with
the increase in number of phases and switching frequency. On the
ripple for the same effective total current ripple. Thus, an increas-
and switching frequencies. The total inductor loss decreases
from one-phase configuration to two-phase configuration, and
then increases again for three-phase configuration. However, the
overall core volume decrease serves as an important tradeoff for
the decreased efficiency.
B. Selection of the Required Switches
In this section, the loss characteristics are analyzed for differ-
ent power modules. The loss calculation for the power modules
is compared based on the datasheet of each module considered
in the comparison. The selected modules are shortlisted based
on their current/voltage ratings, power losses, and package di-
mensions to provide high power density design options. The
characteristics of the selected modules are provided in Table VI.
The parameters of the modules highly depend on the operat-
ing conditions. Half-bridge module is selected for single- and
4148 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021

TABLE VI TABLE VIII

SELECTED MODULE TYPE AND PARAMETERS FOR LOSS COMPARISON COMPARISON OF COMPONENT VOLUMES FOR DIFFERENT CONFIGURATIONS

All the parameters are extracted from the datasheets for a blocking voltage of 800 V.

TABLE VII
COMPARISON OF TOTAL LOSS ESTIMATION

two-phase configurations and is considered as a good package

for robust design and better thermal performance due to the
large surface area of the module. For three-phase configuration,
a six-pack power module is selected as an optimum choice where
the configuration can be designed using only two modules.
Table VII shows the efficiency comparison among all the
configurations using different modules and different switch-
ing frequencies. It is shown that the efficiency of three-phase Fig. 5. Comparison of different SiC-MOSFET-based topologies with a total
SiC MOSFET configuration operating at 20-kHz switching fre- volume less than 180 in3 .
quency is the most efficient solution. However, although higher
switching frequency increases losses for all the configurations,
of output capacitors. The input capacitor is kept the same as
it reduces the inductance and capacitance requirements. This, in
the output capacitor for the capacitor volume calculation, i.e.,
turn, yields a higher volumetric power density design that makes
C1 = C2 . The wire gauges of the inductor windings are selected
it a suitable configuration for the intended application. More-
so that maximum temperature rise is around of 20–30 ◦ C.
over, the reliability of the three-phase configuration is higher,
As a summary, Fig. 5 shows the comparison of the SiC
and the converter is more fault tolerant. Three-phase 100-kHz
MOSFET designs that has a volume less than 180 in3 . As shown,
switching frequency option yields a 98.25% efficiency, which is
the three-phase 100-kHz SiC module design results in 55.6 in3
only ∼1% lower than the most efficient design, i.e., three-phase
volume, a 17% reduction from the second best design, i.e.,
20-kHz SiC design. However, we also need to understand the
three-phase 60-kHz SiC module, only at an efficiency decrease
impact on the total converter size to select the best design option.
of 0.5%. Therefore, for efficient and high volumetric power den-
Table VIII shows the estimated volume of the converter
sity charger, the results conclude that three-phase configuration
based on the volume of main components. The results show Si-
operating at 100 kHz is considered as a proper design solution
IGBT-based configuration in Table VIII(a) and SiC-MOSFET-
for V2V charging application.
configuration in Table VIII(b). The calculated volumes represent
the volume of the switches, inductors, capacitors, and heat sinks,
which form a significant portion of the volume of the design. The IV. SIMULATION RESULTS OF 27-KW V2V CHARGER
gate drivers, control boards, electromagnetic interference filter, For the verification of the required inductance value for dif-
enclosure, and cooling fans are excluded in the comparison. The ferent converter configurations and switching frequencies, we
selected capacitors are film type due to their higher rms current run three simulations modeling the 27-kW charger for the input
capability and lifetime. MKP1848C61090KK2 (10 μF, 900 V) (Vin ) and output (Vout ) voltages, as described in Table IV. The
from Vishay is considered for one-phase and two-phase config- sum of the inductor currents is shown in Fig. 6 for the three
urations and MKP1848C55090JK2 (5 μF, 900 V) is considered interleaved configurations. The mean values of the currents are
for the three-phase configuration. A 10% output current ripple recorded as 67.5 A for the single-phase, 45 A for the two-phase,
is considered as a design constraint for determining the number and 40.5 A for the three-phase configurations. The resulting

Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
UCER et al.: ANALYSIS, DESIGN, AND COMPARISON OF V2V CHARGERS FOR FLEXIBLE GRID INTEGRATION
Fig. 6. Simulation results of the sum of the inductor currents at the rated
power of 27 kW for (a) single-, (b) two-, and (c) three-phase interleaved boost
converters.
ripple currents for each phase are all measured as 3.38 A, which
correspond to 10% of the output current of Iout = Pout ÷ Vout =
27, 000 ÷ 800 = 33.75 A. The simulations confirm that we can
achieve the same current ripple specification at the same power
level with lower inductance values by increasing the number of
phases and/or the switching frequency.
In a longer duration operation, the battery voltage of the
supplier EV decreases and receiver EV increases as charging
continues. If the initial supplier voltage is higher, the converter
operates in the buck mode. However, a transition from buck
mode to boost mode may become necessary as the supplier
voltage drops below the receiver voltage. The problem arises
when the supplier and receiver voltages are very close to each
other (V1 ≈ V2 ). This transition changes the converter dynamics
and causes high voltage and current transients. To overcome this
problem, a smooth transition method can be performed. This
can be achieved by implementing simultaneous switching of the
buck and boost legs of the converter [Sa1 and Sa4 in Fig. 1(a)]
during mode transition, depending on the value of the control
signal [49].
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4149
Fig. 7. Average model results for the voltage transition simulation: SOC,
current, voltage, and power for supplier/receiver vehicles.
To explain longer duration V2V operation and associated
changes that may occur, we developed averaged simulation mod-
els for the V2V converter at a charging power of 27 kW. We mod-
eled a supplier vehicle battery rated at 66.2 kWh and a receiver
vehicle battery at 100 kWh. We initiated the energy sharing with
supplier and receiver EVs at 80% and 20% SOCs, respectively,
while V1 > V2 . The charging current for the receiver EV is set to
64.2 A [constant current (CC) charging] to supply a maximum
power of 27 kW from the supplier at 97% operation efficiency.
The simulation is run to deliver 45 kWh of energy to the receiver
vehicle. Fig. 7 shows the SOC, current, voltage, and power
waveforms of the supplier and receiver vehicles. We can see that
the supplier SOC decreases, whereas the receiver SOC increases
as the CC charging takes place at 64.2 A. The transition happens
around 75 min where the supplier voltage drops below that of
the receiver, and the converter momentarily goes into transition
mode of operation.
V. CLOSED-LOOP CONTROLLER DESIGN
This section develops a design principle of the closed-loop
current controller for the interleaving operation as a basis for the
hardware demonstration presented in Section VI. The operating
conditions are selected based on the available hardware setup
in our laboratory. The reasoning behind doing this is to stay
within the same ratings with the experimental setup explained
in Section VI. The simulations are performed near the operating
output power of 2.5 kW for the single-phase and two-phase
interleaved configurations to verify the proposed ripple current
reduction advantages. The input and output voltages are set
such that Vin = 250 V and Vout = 300 V for boost operation,
and Vin = 300 V and Vout = 250 V for buck operation (these
values are selected based on the input/output capabilities of the
dc power supply and programmable load in our laboratory). The
switching frequency (fs ) is chosen to be 20 kHz (this value is
also decided based on the limitations of the digital controller in
our laboratory). We should also note that we run the computer
simulation and hardware experiment considering a relatively
short amount of charging time where the battery SOCs do not
change considerably. Therefore, this operation does not result in
a transition from buck mode to boost mode. The main purpose
of the simulations and experiment is to show that the energy
transfer can be controlled in either of the operation modes.
4150 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021

Fig. 8. Closed-loop diagram for the current controlled interleaved converter.

For the simulations and experiments, we mainly considered

CC charging for the battery as it can deliver higher amount of
energy per given time at nearly constant power. This operation
is more suitable compared to constant voltage (CV) charging in
terms of charging duration and, thus, more practical as far as
V2V energy sharing is concerned. However, charging can also
take place in the CV mode if the receiver vehicle SOC nears 
IL (s)
70%–80%. The designed controller for the CC model can still Fig. 9. Frequency response curves of d(s)
for multiple-phase configu-
rations for boost operation.
be used as an inner loop for the CV control mode of operation.
This will naturally result in longer charging times but will not
impact the V2V energy transfer.
To control the total supplied current between the two EVs,
a closed-loop average-mode current controller that will track
the sum of all the inductor currents is designed. The overall
closed-loop block diagram is illustrated in Fig. 8. For the single-
phase configuration of each mode (buck/boost), the feedback
variable is the just inductor current itself. For the two-phase
configuration, the feedback is the sum of the inductor currents.
The current controller is designed based on the loop transfer
function of the ratio of the inductor current to the control variable
d (duty ratio), i.e., iL (s)/d(s). The transfer functions for each
mode are derived as follows:
2Vout 1 + s RC
id (s) =
Gboost · 2
(1 − d)2 R 1 + s (1−d)
L
2 R + s (1−d)2
2 LC

Vout 1 + sRC
id (s) =
Gbuck · (4)
dR 1 + s R
L
+ s2 LC
where R is the load resistor value that emulates the behavior
of a battery, d is the duty ratio, L = 260 μH is the inductor,
and C = 100 μF is the output capacitor of the converter. For
the two-phase interleaved configuration, the phase inductors
are selected the same, i.e., L1 =L2 = 260 μH to compare the
ripple reduction. R can be adjusted based on the operating
point of the converter. 2.5 kW of output power approximately
corresponds to an R value of (300 V )2 ÷ 2.5 kW = 36 Ω, and
(250 V )2 ÷ 2.5 kW = 25 Ω based on the operating input and
output voltages for boost and buck operations, respectively. Note
that the control to output transfer functions (4) slightly change as
a new interleaving phase is added to the converter. The frequency
responses of the controlled variable ( IL ) for the single- and
two-phase interleaved configurations are presented in Figs. 9
and 10 for boost and buck modes, respectively. As seen, the
additional phase causes the system pole to shift to a higher
frequency. This means that the controller has to be faster as the
number of phases increases. Therefore, the crossover frequency
of the controlled system needs to be shifted to right to get similar

IL (s)
Fig. 10. Frequency response curves of d(s)
for multiple-phase configu-
rations for buck operation.
closed-loop dynamics in the interleaved configuration. We also
observe that the buck converter requires faster control compared
to the boost converter as it has higher frequency poles.
With the frequency responses in hand, we can design the
closed-loop current controller. First, we need to choose a desired
crossover frequency for each configuration. The phase plots
show that the phase of each loop transfer function for each
mode flattens at −90◦ nearly after 1 kHz. By inspection, we
can choose the crossover frequencies above this point such that
1φ
fxo = 4 kHz and fxo2φ
= 6 kHz considering the switching fre-
quency fsw = 20 kHz. The controller transfer function contains
one zero and two poles, and its transfer function is given in (5).
One of the poles is located at the origin (integrator) providing
a −90◦ phase shift. The locations of the other pole-zero pairs
for a desired phase margin of 60◦ are calculated and listed in

Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
UCER et al.: ANALYSIS, DESIGN, AND COMPARISON OF V2V CHARGERS FOR FLEXIBLE GRID INTEGRATION
TABLE IX
POLE-ZERO LOCATIONS OF THE CONTROLLER TRANSFER FUNCTION
Fig. 11. Simulation results of the sum of the inductor currents for single- and
two-phase interleaved configurations for (a) boost and (b) buck converters.
EXPERIMENTAL RESULTS FOR CURRENT RIPPLE VALUES
Table IX for buck and boost modes
kc (1 + s/wz )
C(s) = . (5)
s (1 + s/wp )
The controllers are designed, tested, and verified in simu-
lations for single- and two-phase configurations and the ripple
results are demonstrated in Fig. 11(a) for boost, and in Fig. 11(b)
for buck mode of operations. The resulting ripple values are
presented in Table VI under Sim. column. We can observe very
close ripple levels in both boost and buck modes, and confirm
that the ripple is reduced with the additional phase.
VI. EXPERIMENTAL TEST RESULTS
In Section II, we presented an analysis for single- and mul-
tiphase interleaved bidirectional buck–boost converters to cal-
culate the worst-case inductance value for an operating point of
27 kW at 20 kHz that covers a wide range of supplier/receiver
battery voltages. Section III presented a comparison study be-
tween Si and SiC based converters for the same operating point
(27 kW) in terms of converter size and efficiency. Section IV
demonstrated simulation results both for the ripple comparison
among three interleaved operations with different inductance
and switching frequencies and the transition operation between
buck and boost modes. Then, in Section V, we presented a
closed-loop current controller design for interleaved converters
and verified the ripple reduction through single- and two-phase
simulations in buck and boost modes. To further support and ver-
ify the analysis and simulation results in the previous sections, an
experimental study is performed in this section to demonstrate
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4151
Fig. 12. V2V charging system test setup.
the effectiveness of the technique and verify the operation of
closed-loop controller design.
A special purpose built back-to-back inverter system and
dSPACE MicroLabBox are used, as illustrated in Fig. 12. The
back-to-back inverter system is composed of a dual three-phase
inverter ready to be interfaced with dSPACE MicroLabBox
development system. This system is rated at 10 kW with the
maximum switching frequency of 20 kHz and a maximum
input/output voltage of 750 VDC. For the three-phase modules,
APTGT50A120T1G IGBTs from Microsemi with SP1 package
is used. As for the passive components, a 260-μH inductor
for each phase and a 100-μF film capacitor for the output are
used. The dSpace MicroLabBox functions as the electronics
control module where all high-level and low-level controllers
are executed. The algorithms used are developed in MATLAB
Simulink, and the execution code is reduced via real-time code
generation. For the emulation of the supplier battery, a 10-kW
MagnaPower XR series power supply is used. For the receiver
battery, a 2.5-kW ALx Series MagnaLOAD dc electronic load is
used via controlling it in the CV mode. The real hardware rack
setup is also shown in Fig. 1(d).
The hardware system is tested for both boost and buck op-
erations for single- and two-phase interleaved typologies. We
were not able to test the three-phase topology with the available
digital controller due to not having access to 120 ◦ phase-shifted
pulsewidth modulation generation. Vin =250 V and Vout =300 V
are chosen as battery voltage levels of supplier and receiver
vehicles for boost mode, and Vin =300 V and Vout =250 V
are selected for buck mode of operation. These voltage levels
are relatively low compared to battery pack voltages of today;
however, they are intentionally set low to have a wide range
over the controlled current without exceeding the power limits
of the electronic load. The switching frequency is set to be
20 kHz for all operations, and Ts =0.05 ms is used for the
closed-loop controller and data acquisition sampling period.
For both operations (boost and buck), the sum of the inductor
currents is controlled for consistency, and to better observe the
impact of additional phases.
Fig. 13(a) and (b) depicts the input and output current and
voltage waveforms for single- and two-phase boost converters,
respectively. The effective switching frequency seen from the
input is doubled due to the additional phase. This results in a
26.31% decrease in the input current ripple for the two-phase de-
sign compared to the single-phase one. Fig. 13(c) and (d) shows
the current and voltage waveforms for single- and two-phase
buck converters, respectively. We observe the same effective
frequency for the buck converter, and the ripple current reduces
4152
Fig. 13. Comparison of (a) single- and (b) two-phase boost mode of operation
results, and (c) single- and (d) two-phase buck mode of operation results (voltage:
100 V/div, current: 5 A/div, and t: 20 µs/div).
Fig. 14. Comparison of measured and reference current to verify closed-loop
control in (a) single-phase and (b) two-phase boost mode.
by 30%. Table VI summarizes the test results and compares them
with the calculation results acquired by the analysis done in Sec-
tion II and the simulation results presented in Section V. There
is a small difference between the experimental and theoretical
results. The actual operating duty ratios can be slightly different
than the calculated ones in the analysis due to nonideal operating
conditions. This results in a slight difference in the calculated
and actual duty ratio of the system. However, the experimental
results are within a reasonable margin of error with the analysis
results. Furthermore, the expected ripple reduction is attained.
To test the performance of the closed-loop controller designed
in Section V, we commanded different current values as our
reference. The resulting currents are recorded at the sampling
frequency of 20 kHz and demonstrated along with the reference
current in Fig. 14(a) and (b) for the boost single- and two-phase,
and in Fig. 15(a) and (b) for the buck single- and two-phase
converters. The transient performance of the controllers is also
demonstrated within 1-s interval in Fig. 16(a) and (b) for the
boost single- and two-phase, and in Fig. 17(a) and (b) for the
buck single- and two-phase converters. The small overshoots
observed especially in the two-phase results are caused mainly
due to not having enough sampling speed in the digital controller
and the filtering delays that are present in the feedback path.
Nevertheless, the controller works stable and the system follows
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021
Fig. 15. Comparison of measured and reference current to verify closed-loop
control in (a) single-phase and (b) two-phase buck mode.
Fig. 16. Transient response of the controller in (a) single-phase and (b) two-
phase boost mode.
Fig. 17. Transient response of the controller in (a) single-phase and (b) two-
phase buck mode.
the commanded currents. We see that the experimental results
verify our theoretical and simulation results in that the current
ripple is being reduced by each additional phase.
VII. CONCLUSION
The study carried out in this article envisions the viability
of a V2V energy sharing using compact and flexible charging
infrastructure. Such infrastructure will help facilitate widespread
penetration of EVs into the market reducing the reliance on
expensive and nonubiquitous DCFC infrastructure. The overall
merit of this study lies in the design of a V2V charger to
be used in CSN. The study presents the design of a compact
bidirectional two-way buck–boost dc–dc converter to provide
a proof of concept for energy transfer between two different
battery packs of any typical voltage level. We analyzed three
candidate options to transfer energy between two wide battery
voltage levels and compared their advantages. A significant
improvement in inductor ripple current is observed when multi-
phase bidirectional dc–dc converters are used as opposed to the
single-phase counterpart at the expense of increased complexity.
To quantitatively compare the design topologies for Si and SiC
UCER et al.: ANALYSIS, DESIGN, AND COMPARISON OF V2V CHARGERS FOR FLEXIBLE GRID INTEGRATION 4153

based converters, we carried out a design comparison using effi- [18] “Hyundai Kona to get V2V charging.” 2019. [Online]. Available: https:
cient and compact over-the-shelf power modules and magnetic //gomechanic.in/blog/hyundai-kona-to-get-v2v-charging/
[19] “Plugged in: How Americans charge their electric vehicles,” 2015.
cores. The comparison shows how multiphase interleaved SiC Accessed: Apr. 18, 2019. [Online]. Available: https://avt.inl.gov/sites/
converter topologies enable high switching frequency operation. default/files/pdf/arra/PluggedInSummaryReport.pdf
This way the inductance requirements and the converter size [20] E. Bulut and M. C. Kisacikoglu, “Mitigating range anxiety via vehicle-
to-vehicle social charging system,” in Proc. IEEE Veh. Technol. Conf.,
significantly reduce while maintaining high system efficiency. Jun. 2017, pp. 1–5.
The proposed work aims to enable significant spread of V2V [21] E. Bulut, M. C. Kisacikoglu, and K. Akkaya, “Spatio-temporal non-
technology to help the flexible integration of EVs into the power intrusive direct V2V charge sharing coordination,” IEEE Trans. Veh.
Technol., vol. 68, no. 10, pp. 9385–9398, Oct. 2019.
grid. This allows to mitigate the EV charging loading impact on [22] R. Zhang, X. Cheng, and L. Yang, “Flexible energy management protocol
the grid and provides indirect grid services. for cooperative EV-to-EV charging,” IEEE Trans. Intell. Transp. Syst.,
vol. 20, no. 1, pp. 172–184, Jan. 2019.
[23] F. Yucel, K. Akkaya, and E. Bulut, “Efficient and privacy preserving
supplier matching for electric vehicle charging,” Ad Hoc Netw., vol. 90,
REFERENCES 2019, Art. no. 101730.
[24] B. Roberts, K. Akkaya, E. Bulut, and M. Kisacikoglu, “An authenti-
[1] E. Ucer et al., “A flexible V2V charger as a new layer of vehicle-grid
cation framework for electric vehicle-to-electric vehicle charging appli-
integration framework,” in Proc. IEEE Transp. Electrific. Conf. Expo.,
cations,” in Proc. IEEE MASS REU Res. Netw. Syst. Workshop, 2017,
2019, pp. 1–7.
pp. 565–569.
[2] E. Veldman and R. A. Verzijlbergh, “Distribution grid impacts of smart
[25] S. Dusmez, A. Hasanzadeh, and A. Khaligh, “Comparative analy-
electric vehicle charging from different perspectives,” IEEE Trans. Smart
sis of bidirectional three-level dc-dc converter for automotive appli-
Grid, vol. 6, no. 1, pp. 333–342, Jan. 2015.
cations,” IEEE Trans. Ind. Electron., vol. 62, no. 5, pp. 3305–3315,
[3] L. P. Fernandez, T. G. S. Roman, R. Cossent, C. M. Domingo, and
May 2015.
P. Frias, “Assessment of the impact of plug-in electric vehicles on dis-
[26] O. Garcia, P. Zumel, A. de Castro, and A. Cobos, “Automotive dc-dc
tribution networks,” IEEE Trans. Power Syst., vol. 26, no. 1, pp. 206–213,
bidirectional converter made with many interleaved buck stages,” IEEE
Feb. 2011.
Trans. Power Electron., vol. 21, no. 3, pp. 578–586, May 2006.
[4] S. Shafiee, M. Fotuhi-Firuzabad, and M. Rastegar, “Investigating the
[27] T. J. C. Sousa, V. Monteiro, J. C. A. Fernandes, C. Couto, A. A. N.
impacts of plug-in hybrid electric vehicles on power distribution systems,”
Melendez, and J. L. Afonso, “New perspectives for vehicle-to-vehicle
IEEE Trans. Smart Grid, vol. 4, no. 3, pp. 1351–1360, Sep. 2013.
(V2V) power transfer,” in Proc. IEEE Ind. Electron. Soc. Annu. Conf.,
[5] F. Erden, M. C. Kisacikoglu, and O. H. Gurec, “Examination of EV-grid
Oct. 2018, pp. 5183–5188.
integration using real driving and transformer loading data,” in Proc. Int.
[28] M. O. Badawy et al., “Design and implementation of a 75-kW mobile
Conf. Elect. Electron. Eng., 2015, pp. 364–368.
charging system for electric vehicles,” IEEE Trans. Ind. Appl., vol. 52,
[6] E. Sortomme, M. M. Hindi, S. J. MacPherson, and S. Venkata, “Coordi-
no. 1, pp. 369–377, Jan./Feb. 2016.
nated charging of plug-in hybrid electric vehicles to minimize distribution
[29] M. Nasr, K. Gupta, C. da Silva, C. H. Amon, and O. Trescases, “SiC
system losses,” IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 198–205,
based on-board EV power-hub with high-efficiency DC transfer mode
Mar. 2011.
through AC port for vehicle-to-vehicle charging,” in Proc. IEEE Appl.
[7] N. Leemput, F. Geth, J. Van Roy, A. Delnooz, J. Buscher, and J. Driesen,
Power Electron. Conf. Expo., 2018, pp. 1–7.
“Impact of electric vehicle on-board single-phase charging strategies
[30] S. Zeljkovic, R. Vuletic, A. Miller, and A. Denais, “A three phase bidirec-
on a Flemish residential grid,” IEEE Trans. Smart Grid, vol. 5, no. 4,
tional V2G interface converter based on SiC JFETs,” in Proc. 17th Eur.
pp. 1815–1822, Jul. 2014.
Conf. Power Electron. Appl., Sep. 2015, pp. 1–10.
[8] K. Clement-Nyns, E. Haesen, and J. Driesen, “The impact of charging
[31] A. Hilal and B. Cougo, “Optimal inductor design and material selection
plug-in hybrid electric vehicles on a residential distribution grid,” IEEE
for high power density inverters used in aircraft applications,” in Proc. Int.
Trans. Power Syst., vol. 25, no. 1, pp. 371–380, Feb. 2010.
Conf. Elect. Syst. Aircr., Railway, Ship Propulsion Road Veh. Int. Trans.
[9] O. Sundstrom and C. Binding, “Flexible charging optimization for electric
Electrific. Conf., Nov. 2016, pp. 1–6.
vehicles considering distribution grid constraints,” IEEE Trans. Smart
[32] M. Gjelaj, C. Træholt, S. Hashemi, and P. Andersen, “Optimal design of
Grid, vol. 3, no. 1, pp. 26–37, Mar. 2012.
DC fast-charging stations for EVs in low voltage grids,” in Proc. IEEE
[10] N. Leemput, F. Geth, J. Van Roy, P. Olivella-Rosell, A. Sumper, and
Transp. Electrific. Conf. Expo., Jun. 2017, pp. 684–689.
J. Driesen, “MV and LV residential grid impact of combined slow and
[33] L. Albiol-Tendillo, E. Vidal-Idiarte, J. Maixé-Altés, J. M. Bosque-
fast charging of electric vehicles,” Energies, vol. 8, no. 3, pp. 1760–1783,
Moncusí, and H. Valderrama-Blaví, “Design and control of a bidirectional
2015.
DC/DC converter for an electric vehicle,” in Proc. 15th Int. Power Electron.
[11] “Global EV outlook: Beyond one million electric cars,” Tech. Rep.,
Motion Control Conf., Sep. 2012, pp. LS4d.2-1–LS4d.2-5.
2016. Accessed: Apr. 18, 2019. [Online]. Available: https://www.iea.org/
[34] F. M. Almasoudi, K. S. Alatawi, and M. Matin, “Design of isolated
publications/freepublications/publication/Global_EV_Outlook_2016.
interleaved boost DC-DC converter based on SiC power devices for
pdf
microinverter applications,” in Proc. North Amer. Power Symp., Sep. 2016,
[12] W. Kempton and S. Letendre, “Electric vehicles as a new power source
pp. 1–6.
for electric utilities,” Trans. Res., vol. 2, no. 3, pp. 157–175, Sep. 1997.
[35] G. R. Chandra Mouli, J. H. Schijffelen, P. Bauer, and M. Zeman, “Design
[13] “Charging forward: The state of EV charging in 2016,” 2016. Ac-
and comparison of a 10-kW interleaved boost converter for PV application
cessed: Apr. 18, 2019. [Online]. Available: https://www.chargepoint.com/
using Si and SiC devices,” IEEE J. Emerg. Sel. Topics Power Electron.,
charging-forward-2016
vol. 5, no. 2, pp. 610–623, Jun. 2017.
[14] N. Nigro and M. Frades, “Business models for financially
[36] C. N. Ho, H. Breuninger, S. Pettersson, G. Escobar, and F. Canales, “A
sustainable EV charging networks,” Center for Climate and
comparative performance study of an interleaved boost converter using
Energy Solutions, Tech. Rep., 2015. Accessed: Apr. 18, 2019.
commercial Si and SiC diodes for PV applications,” IEEE Trans. Power
[Online]. Available: http://leg.wa.gov/JTC/Documents/Studies/EV/
Electron., vol. 28, no. 1, pp. 289–299, Jan. 2013.
FinalReport_EVChargingNetworksWEB.pdf
[37] S. Öztürk, “Design of three phase interleaved DC/DC boost converter with
[15] J. Leeder, “Ontario’s electric-vehicle charging network hits speed
all SiC semiconductors for electric vehicle applications,” in Proc. 10th Int.
bumps,” 2017. Accessed: Apr. 18, 2019. [Online]. Available: https:
Conf. Elect. Electron. Eng., Nov. 2017, pp. 355–359.
//www.theglobeandmail.com/news/toronto/ontarios-electric-vehicle-
[38] R. Mayer, M. Berrehil El Kattel, M. D. Possamai, C. Bruning, and
charging-network-hits-speed-bumps/article34441795
S. V. Garcia Oliveira, “Analysis of a multi-phase interleaved bidirectional
[16] “Rivian confirms V2V charging, auxiliary batteries.” 2019. [Online].
DC/DC power converter with coupled inductor,” in Proc. Brazilian Power
Available: https://www.autoblog.com/2019/06/08/rivian-confirms-v2v-
Electron. Conf., Nov. 2017, pp. 1–6.
charging-auxiliary-batteries/slide-1282136
[39] Z. Kan, P. Li, R. Yuan, and C. Zhang, “Interleaved three-level bi-directional
[17] “Hyundai introduces vehicle to vehicle charging facility.” 2019. [Online].
DC-DC converter and power flow control,” in Proc. 3rd Int. Conf. Intell.
Available: https://www.autocarindia.com/car-news/hyundai-introduces-
Green Building Smart Grid, Apr. 2018, pp. 1–4.
vehicle-to-vehicle-charging-facility-415144

Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
4154 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 57, NO. 4, JULY/AUGUST 2021
[40] T. Kang, C. Kim, Y. Suh, H. Park, B. Kang, and D. Kim, “A design and
control of bi-directional non-isolated DC-DC converter for rapid electric
vehicle charging system,” in Proc. IEEE Appl. Power Electron. Conf.
Expo., Feb. 2012, pp. 14–21.
[41] B. Vural, S. Dusmez, M. Uzunoglu, E. Ugur, and B. Akin, “Fuel consump-
tion comparison of different battery/ultracapacitor hybridization topolo-
gies for fuel-cell vehicles on a test bench,” IEEE J. Emerg. Sel. Topics
Power Electron., vol. 2, no. 3, pp. 552–561, Sep. 2014.
[42] Y. Du, X. Zhou, S. Bai, S. Lukic, and A. Huang, “Review of non-isolated
bi-directional dc-dc converters for plug-in hybrid electric vehicle charge
station application at municipal parking decks,” in Proc. IEEE Appl. Power
Electron. Conf. Expo., Feb. 2010, pp. 1145–1151.
[43] C. Chang and M. A. Knights, “Interleaving technique in distributed power
conversion systems,” IEEE Trans. Circuits Syst. I, Fundam. Theory Appl.,
vol. 42, no. 5, pp. 245–251, May 1995.
[44] M. Gerber, J. A. Ferreira, I. W. Hofsajer, and N. Seliger, “Interleaving
optimization in synchronous rectified dc/dc converters,” in Proc. IEEE
Power Electron. Specialists Conf., Jun. 2004, vol. 6, pp. 4655–4661.
Authorized licensed use limited to: The University of Akron Libraries. Downloaded on July 20,2021 at 13:30:01 UTC from IEEE Xplore. Restrictions apply.
[45] D. Baba, “Under the hood of a multiphase synchronous rectified boost
converter,” Texas Instruments, Tech. Rep., 2014. Accessed: Apr. 18, 2019.
[Online]. Available: https://www.ti.com/seclit/wp/slup323/slup323.pdf
[46] C. Parisi, “Multiphase buck design from start to finish (Part 1),” Texas In-
struments, Tech. Rep., 2017. Accessed: Apr. 18, 2019. [Online]. Available:
http://www.ti.com/lit/an/slva882/slva882.pdf
[47] “Express 100 DC commercial charging stations.” 2018. [Online]. Avail-
able: https://www.chargepoint.com/files/datasheets/ds-cpe100.pdf
[48] A. Anthon, Z. Zhang, M. A. E. Andersen, D. G. Holmes, B. McGrath, and
C. A. Teixeira, “The benefits of SiC mosfets in a T-type inverter for grid-tie
applications,” IEEE Trans. Power Electron., vol. 32, no. 4, pp. 2808–2821,
Apr. 2017.
[49] L. Callegaro, M. Ciobotaru, D. J. Pagano, E. Turano, and J. E. Fletcher,
“A simple smooth transition technique for the noninverting buck-boost
converter,” IEEE Trans. Power Electron., vol. 33, no. 6, pp. 4906–4915,
Jun. 2018.