Chapter 6

Reducing grid dependency of EV charging using

renewable and storage systems
Viet Thang Tran1, Md Rabiul Islam1, Danny Sutanto1, and
Kashem M Muttaqi1

Electric vehicles (EVs) are increasingly deployed and may become the main private
and public transportation mode in many countries in the future. This is the result
of the clean energy programs promoted by many governments all over the world
to deal with climate change, where incentives are offered for low- or zero-­carbon
emission vehicles to reduce the greenhouse gas emission from the road transport
sector. Such incentives encourage automakers to increase EV production capacity.
Following widespread deployment of EVs, it is envisaged that EV charging may
become a dominant load in the power grid due to the high power required to charge
the EV battery, particularly when EVs can be charged whenever they are parked at
home or in parking lots. In such a scenario, the grid may experience an unpredict-
ably large variation in the load demand that can lead to various technical issues, such
as voltage regulation, harmonic contamination, and frequency variations. These will
bring significant challenges to the operation and management of the power utili-
ties. Therefore, to keep up with the increasing penetration of EVs, the distribution
grid needs to be upgraded to increase its capacity to meet the new demand from
EV charging. Unfortunately, such upgrades will take time and will slow down the
deployment of EVs.
This chapter presents the solution of using solar photovoltaic (PV) and an energy
storage unit to mitigate the effects of the high impact of EVs in all levels of charging
from Level 1 to Level 3 of the EV charging system. The chapter is organized as fol-
lows: Section 6.1 presents an overview of the EV charging configurations. Section 6.2
presents the solution for mitigating the effects of high penetration of EVs with the
Level 1 and Level 2 charging system. Section 6.3 proposes the solution for mitigat-
ing the effects of EVs with the Level 3 charging system. Conclusions are drawn in
Section 6.4.

School of Electrical, Computer and Telecommunications Engineering, University of Wollongong,

1

Australia
164 Cable based and wireless charging systems for electric vehicles

Figure 6.1   General configuration of an EV charging system

6.1 EV charging system

EV charging systems are increasingly being installed in public areas to facilitate the
widespread use of EVs in the public domain [1, 2]. However, the high penetration of
EVs can bring undesirable impacts on the distribution grid [3–7]. These may bring
unexpected challenges for the power utilities for the operation and management of
the grid system. Therefore, EV charging systems are continuously being improved
to keep up with the increasing penetration of EVs. According to the charging stand-
ard [8], the maximum power at Level 1 and Level 2 charging is 19.2 kW. However,
the power rating of a Level 1 portable charging cable that can be plugged directly
into an electric power point is normally far lower than the standard maximum power
allowable at Level 1, which is about 1–2.4 kW, which can comply with the maxi-
mum current rating of standard power points (10 A) in residential homes, offices, or
shopping centers. Level 3 charging refers to a dc charging system that has a charging
power rating ranging from 30 kW up to hundreds of kW.

6.1.1 EV charger topologies

In general, the EV chargers can be categorized as on-­board and off-­board chargers
corresponding to Levels 1, 2, and 3 of the charging standards [9–19]. Figure 6.1
shows a general configuration of an EV charging system.
The on-­board charger comprises the power converters shown as (3) and (4) in
Figure 6.1, which should have high power density and high efficiency. The con-
verter denoted as (3) in Figure 6.1 is normally a power factor correction (PFC)
rectifier, which consists of an uncontrolled full-­bridge rectifier and a power boost
converter. The converter denoted as (4) in Figure 6.1 can be a boost converter or an
isolated dc-­dc converter such as the dual active bridge (DAB) converter or H-­bridge
inductor-­capacitor-­capacitor resonant converter [10]. Boost type converters are nor-
mally used on account of their simplicity in control, high efficiency, and capabil-
ity of being easily modified into a bidirectional converter by replacing a switching
 Reducing grid dependency of EV charging 165

device in place of a power diode [9, 15]. However, the lack of isolation in them
results in difficulty to achieve high power transfer. To solve this issue, a DAB is a
suitable converter as it has a high-­frequency transformer acting as an isolator. This
makes it able to achieve high power and bidirectional power flow. However, the
control of DAB is more complicated because of the required high-­computational
algorithm [18, 19].
To control the charging procedure, electric vehicle service equipment (EVSE)
is needed. The EVSE will have the function of turning on/off the ac power from
the grid when the electric vehicle battery (EVB) gets fully charged and provide a
charging fee or bill. The dc power in the dc link can also be used to supply the elec-
tronic loads in the car through the dc-­dc converter denoted as (1) in Figure 6.1. The
off-­board charger is a Level 3 dc fast charger and is located in the charging station.
The converter of the off-­board charger, denoted as (5) in Figure 6.1, is a high-­power
dc-­dc converter, which can be a multilevel converter or a cascaded converter. The
converter denoted as (6) in Figure 6.1 is a three-­phase PFC rectifier or a bidirectional
converter that not only can charge the EV batteries but also can send power back to
the grid.

6.1.2 EV charging/discharging strategies

The EV battery can be charged from or discharged to the distribution grid. Depending
on the specific application, an EV charger can be designed to be either unidirectional
or bidirectional. The current commercial EV charger is designed with a unidirec-
tional converter. This means that EVs can be charged from the grid or renewable
sources. The charging strategies for a unidirectional EV charger can be uncoordi-
nated or coordinated. The uncoordinated charging strategy means that the EV can be
charged at any time when it is plugged into the grid. This strategy has the potential to
stress the distribution grid if a fleet of EVs is allowed to be charged at the same time.
To mitigate the effects of EV charging on the grid, the coordinated charging strategy
is proposed by researchers [20–29]. Several coordinated charging strategies have
been proposed recently, such as the time-­of-­use-­based charging [20], the real-­time
smart load management [22], and the distributed resources-­based charging [23–29].

6.2 Integration of EV charging-home solar PV system

The integration of Level 1 and Level 2 EV charging into the home solar PV (EVC-­
HSP) system is shown in Figure 6.2. The power transfer between the HSP system
and EV battery is implemented based on a power conditioning system (PCS), which
consists of three power converters sharing a dc bus. The charging dc connector is
based on the CHAdeMO standard as it allows EVB to be interfaced with the dc
bus of the HSP system via a power dc-­dc converter named EVB converter. The PV
converter is deployed by a power boost converter to track the maximum power point
(MPP) of the HSP system. The grid converter (GC) is implemented by a full-­bridge
converter to interface with the grid. It is noted that a conventional HSP system con-
sists of a PV converter and GC. In this configuration, the GC is a unidirectional
166 Cable based and wireless charging systems for electric vehicles

Figure 6.2   Configuration of the EVC-­HSP system

converter that only injects the generated HSP power into the grid. However, in the
ECV-­HSP configuration, the GC is a bidirectional converter where the power can
be transferred back and forth from the grid. By adding an EVB converter into the
dc bus of the conventional HSP system, the ECV-­HSP system is formed. The EVB
converter is a unidirectional converter or bidirectional converter. If the EVB gets
charged only from the HSP and grid, it can be designated as a unidirectional Boost
converter. If the EVB has a function of supporting the grid, it can be designated
as a bidirectional Buck/Boost converter. The output of the EVB converter is con-
nected to the CHAdeMO connector. The mechanism of the handshaking signals of
the CHAdeMO is presented in Figure 6.3. A control board is used to control the
power flow of the EVC-­HSP system. With this control mechanism, the effective
power management of the EVC-­HSP is obtained.

6.2.1 Operation modes of EVC-HSP system

The main idea for the EVC-­HSP system is to utilize the HSP system to efficiently
charge the EVB while still supporting the loads at home or grid in the case of EV
traveling. This idea not only helps mitigate the effects of the high penetration of EVs
but also helps increase the HSP system implementation. To achieve this, efficient
operation modes of the EVC-­HSP should be defined clearly. It has been well estab-
lished that the HSP system will produce the highest energy around noon. The daily
HSP power curve achieves a peak around 10 a.m. to 3 p.m. On the other hand, the
home load demand curve often referred to as the “Duck curve”, has the peak load
around 5 p.m. to 9 p.m., with the lowest point of the curve occurring around noon.
This reflects the basic idea that the EVB should be charged around noon and may
discharge to support the home load and grid around the peak time from 5 p.m. to
9 p.m. The EVB may be charged from the grid at night when the electricity price is
low. This also helps the grid become stable as of balancing the generation and power
consumption from the grid. The effective operating modes of the EVC-­HSP system
can be defined as:
 Reducing grid dependency of EV charging 167

Figure 6.3   Mechanism handshaking signals of the CHAdeMo connector

Mode I: The EV is on the road and the home load is low. This mode may happen
at any time during the daytime when people drive the EVs for doing their day-­to-­day
works. The generated HSP power directly injected into the grid is given by (6.1),
where ‍ ‍is the efficiency of the PCS, Pg is the grid power, and Ppv is the PV power:
‍ Pg = Ppv ‍ (6.1)
Mode II: The EV is charged from the HSP power. Depending on the pre-­defined
ref
maximum charging power, the EVB charging power reference ‍PEVB ‍ is calculated
according to (6.2) with the constraints of the maximum EVB state-­of-­charge
(SoCmax).
8̂  
1 max
ˆ
<   Ppv if Ppv k  PEVB , SoC  SoCmax
ref
PEVB = Pmax if Ppv > Pmax
EVB , SoC  SoCmax (6.2)
ˆ EVB
:̂ 0 if SoC > SoCmax
‍ ‍
Mode III: The EVB is used to support the home load demand at peak time. This
not only helps reduce the stress on the grid at the peak demand but also helps save
the electricity bills for the EV owners. The EVB discharging power reference is
calculated based on (6.3), where PL is the home load power with the minimum EVB
state-­of-­charge (SoCmin).
168 Cable based and wireless charging systems for electric vehicles

(a)

(b)

(c)

Figure 6.4  Detailed control system of the EVC-­HSP system: (a) PV converter

control, (b) GC control, and (c) EVB converter control

ref
‍ PEVB = PL  Ppv , SoC  SoC min‍ (6.3)
req
Mode IV: The EVB is used to support the grid with the required power ‍Pg .‍ The
EVB discharging power reference is calculated based on (7.4).
ref
‍ PEVB = PL + Preq
g , SoC  SoCRes ‍ (6.4)
where SoCRes is the reserved energy of the EVB. This mode can be achieved through
the request from the grid operator, and it is a smart function of the smart grid where
multi-­sources are interconnected and cooperated.

6.2.2 Control strategy of EVC-HSP system

The power converter of the EVC-­HSP system is controlled to achieve all operation
modes. Each converter has its specific control structure as depicts in Figure 6.4. The
output voltage of the HSP is tracked for the MPP of the HSP system; the HSP volt-
age reference is found by the perturbation and observation algorithm, and then a PI
controller is designed to track the HSP voltage reference.
The output control signal dpv of the EVB converter is expressed by (6.5), where
k‍ ppv ‍ and ‍kipv ‍ are proportional and integral parameters of the PI0 controller, respec-
tively. V ‍ pv ‍and V ‍ pv ‍are actual and reference voltages of the HSP system.
!
p
kipv  ref 
dpv = kpv + Vpv  Vpv (6.5)
s
‍ ‍
 Reducing grid dependency of EV charging 169

The grid-­connected converter is controlled through the dc bus Vdc to achieve a bidi-
rectional power flow. The control for the GC is shown in Figure 6.4(b). The control
signals of the GC, dvdc, and dig, respectively, are calculated according to (6.6) and
ref ref
(6.7), where V ‍ dc ‍ is the grid
‍ dc ‍ is the voltage reference of the dc bus voltage and V
current reference.
 
p kivdc  ref 
dvdc = kvdc + Vdc  Vdc (6.6)
‍ s ‍
!
p kiig  ref 
dig = kig + Ig  I g (6.7)
s
‍ ‍
The EVB is charged and discharged through the EVB converter. The control of the
EVB converter is shown in Figure 6.4(c), where the constant current control is used
for both charging and discharging processes. The current controller is following
(6.8), where dcc is the control signal, ‍kpcc ‍and ‍kicc ‍are PI3 controller parameters, and ‍kicc ‍
is the EVB charging/discharging current reference.
 
p kicc  ref 
dcc = kcc + IEVB  IEVB (6.8)
‍ s ‍
The constant voltage (CV) control is used when the EVB reaches its gassing voltage
and the EVB is kept to the EVB CV until the EVB is fully charged. The control vari-
able dvc of the voltage controller is the EVB current reference as given according to
ref
(6.9), where ‍kpvc ‍and ‍kivc ‍are PI4 parameters, respectively, and V
‍ EVB ‍is the EVB voltage
reference.
 
p kivc  ref 
dvc = kvc + VEVB  VEVB (6.9)
‍ s ‍

6.2.3 Simulation results of EVC-HSP system

To validate the performance of the proposed EVC-­HSP system and its control strat-
egy, a detailed switching model of the entire EVC-­HSP is simulated in MATLAB®/
Simulink environment. The irradiance profile is selected to ensure that the EVC-­
HSP experiences all the operation modes depicted in Figures 6.5 and 6.6.
The green line is the PV power with a maximum power of 5 kW. The PV power
has been varied through all the operation modes. All the energy generated by the
PV system is utilized for charging to the EVB at the period from t = 0 s to t = 1 s
with the maximum charging current. During the period from t = 1 s to t = 2 s, the
PV irradiance is decreased; however, the EVB is nearing its fully charged condition,
and hence it is required to limit the charging current. Therefore, the excess generated
power from the HSP system will be injected into the grid. During the period from
t = 2 s to t = 3 s, the EVB is fully charged, all the HSP power is injected into the
grid. At the peak load, the EVB will support the load as it will be discharging its
energy in the period from t = 3 s to t = 4 s, after discharging its energy and when
the load demand is low, the EVB starts being charged as shown in the period from
170 Cable based and wireless charging systems for electric vehicles

Figure 6.5   Simulation results of the EVC-­HSP system in four operation modes

t = 4 s to t = 5 s. During the period from t = 5 s to t = 6 s, the EVB supports the grid

during the peak load hours. The EVB power is shown in Figure 6.6.
The simulation results show that with the efficient energy management of the
EVB and PV power, the EV owners will have the benefit of minimizing the elec-
tricity cost and the distribution grid can handle the increasingly high penetration of
the PV power generation and high penetration of EVs as well. Figure 6.7 shows the
state-­of-­charge (SoC) of the EVB through the different operation modes. The grid
voltage and current waveforms of the operation modes are shown in Figure 6.8.

6.2.4 Experimental results of EVC-HSP system

A scaled laboratory prototype is built to test and debug the performance of the EVC-­
HSP system depicted in Figure 6.9. The test platform comprises an emulated 30 kW
from the solar PV TerraSas emulator, a 10-­kWh lead-­acid battery, which is consid-
ered as the EV battery. Moreover, the power converters based on a 5-­kW insulated-­
gate bipolar transistor (IGBT) module from SEMIKRON were also used. The con-
trol system is implemented digitally using a programmable platform from National
Instruments called general-­purpose inverter controller board. The maximum power
generated by the HSP system is 275 W at the MPP voltage of 100 V. The dc-­link

Figure 6.6   Simulation results of the EVB power in four operation modes
 Reducing grid dependency of EV charging 171

Figure 6.7   SoC, current, and voltage of the EVB in operation modes

voltage is maintained at 220 V with the grid voltage is set as 127 V. Two resistive
loads of 60 Ω and 110 Ω, respectively, are considered as the home loads.
Figure 6.10 depicts the performance of the EVC-­HSP system through the charg-
ing and the discharging procedure of the EV battery. Period (a) shows the charging
of the EVB from the PV power and the period (b) shows the discharging of the EVB
when it supports the local load. The EVB is charged in period (c) and it is discharged
in period (d) when there is no PV power.

6.2.5 Summary designing of an EVC-HSP system

An EVC-­HSP system for the integration of the HSP system and EV charging is
presented. The energy management system successfully controls the power balance
between the HSP source, EVB energy, the home load, and the grid demand to ensure
the reduced dependency of the EVB charging power from the power grid. This will
help to mitigate the uncertainty effects caused by the EVB charging. It is envisaged
that the proposed system with its control algorithm will have a great potential to
increase the deployment of EVs.

Figure 6.8  Grid voltage and current of the EVC-­HSP system in four operation
modes
172 Cable based and wireless charging systems for electric vehicles

Figure 6.9  The detailed experimental setup of the laboratory scale of the EVC-­
HSP system

6.3 Level 3 – fast-charging infrastructure with solar PV and

energy storage
To mitigate the effects of the EV fast-­charging by reducing the dependence of the
EV fast-­charging on the grid, a fast-­charging infrastructure (FCI) that can be inte-
grated with an energy buffer unit (EBU) and a PV system is proposed as shown in
Figure 6.11. The proposed FCI not only can help mitigate the effects of the high
penetration of PV and EV into the distribution grid, but it can also play a key role in
the development of the future smart grid. The four vital components of the FCI are
grid converter, PV converter, EBU converter, and EVB converter. All these power

Figure 6.10   Experimental results of the EVC-­HSP system

 Reducing grid dependency of EV charging 173

Figure 6.11   Infrastructure of a future EV fast-­charging station

converters will share the same dc power bus. The number of EVs that can be charged
concurrently depends on the number of EVB converters connected to the dc bus.
The power flow in the proposed FCI is shown in Figure 6.11.
The PV power (PPV) will flow into EBU, EVB, and the distribution grid. The
EBU power (PEBU), EVB power (PEVB), and grid power (Pg) will flow back and forth
between EBU, EVB, and the distribution grid because the EBU converter, EVB con-
verter, and GC are bidirectional. As the power rating per charge ranges from several
dozen kilowatts up to hundreds of kilowatts, if a few EVs are charged concurrently,
the FCI power rating will be a few hundred kilowatts. The power rating is consid-
ered as the design criteria for FCI. To achieve reliable operation in any scenario,
the power rating of the GC should be designed in such a way that it can handle the
power flow in the FCI. The power rating of the GC depends on the power rating of
the PV converter, EBU converter, and EVB converter, and it can be designed based
on (6.10) or (6.11).
P
i=n
Pmax max max
GC = PPV + PEBU + Pmax
EVB (6.10)
‍ ( i=1 ‍ )
P
i=n
Pmax max max
GC = Max PPV , PEBU , Pmax
EVB (6.11)
‍ i=1
‍
where ‍Pmax max max max
GC , ‍‍PPV , ‍‍PEBU , ‍and ‍PEVB ‍are maximum power that GC, PV converter,
EBU converter, and EVB converter can deliver, and n is the number of EVB
converters.
174 Cable based and wireless charging systems for electric vehicles

Figure 6.12   Power flow in the FCI

If the power rating of the GC is designed based on (6.10), the GC will be

capable of operating for all types of power transfer between PV, EVB, EBU,
and the distribution grid. In the worst case, during a blackout or peak demand,
the power delivered to the distribution grid or local load will be the aggrega-
tion of the PV power, EVB power, and EBU power.
In the GC power rating designed based on (6.11), it is assumed that not all
distributed sources can simultaneously inject power into the grid at the same
time. The grid power rating can then be equal to the maximum power of the
PV, EBU, and EVB.
The power flow in the FCI is shown in Figure 6.12. There are four power
nodes: A, B, C, and D, which are defined as the aggregated power of each
node. The positive sign (+) means the power flows into the node and the nega-
tive sign (−) means the power flows out from the node. The power at each node
is expressed as (6.12).
8̂
ˆ
ˆ PA = PPV + ˛0 PAg
ˆ
ˆ
ˆ
ˆ B Ch
< PB = PA + ˛1 Pg + ˛2 PEBU
PC = PB + ˛3 PC Dis
g + ˛4 PEBU (6.12)
ˆ
ˆ
ˆ Ch
ˆ PD = PC + ˛5 PEVB
ˆ
ˆ
:̂ P = P + ˛ PE + ˛ PDis
‍ L D 6 g 7 EBU
‍
The coefficients αi (i=0:7) can be −1 or 0 based on the defined power modes. Nine
operation modes are defined in such a way that EVB is preferentially charged from
the PV and EBU power.
Mode 1: The FCI has no EV that needs to be charged, the generated PV power
will be used to charge the EBU, the power flow at nodes A and B can be expressed
as (6.13).
8̂
ˆ
< PA = PPV ˛0 = 0
P = 0 (6.13)
ˆ B
:̂ P  PCh = 0 ˛1 = 0; ˛2 = 1
‍ A EBU ‍
Mode 2: The EBU is fully charged, no EV needs to be charged. The generated PV
power will be injected into the grid, and the power flow at the power nodes can be
expressed as in (6.14).
 Reducing grid dependency of EV charging 175
(
PPV  PAg = 0 ˛0 = 1
(6.14)
‍ PA = 0, PB = 0, PC = 0, PD = 0 ‍
Mode 3: The EBU is fully charged, the EVB will be charged from the PV power
only. The power at the power nodes can be expressed by (6.15).
8̂
ˆ PA = PPV ˛0 = 0
ˆ
ˆ
< P =P ˛1 = 0; ˛2 = 0
B A
(6.15)
ˆ
ˆ PC = PB
ˆ ˛3 = 0; ˛4 = 0
:̂
‍ PC  PCh
EVB ˛5 = 1 ‍
Mode 4: PV power is insufficient, the EVB will be charged from the generated PV
power and EBU power. In this case, EBU operates under the discharging mode. The
power flow at the power nodes can be expressed by (6.16).
8̂
ˆ PA = PPV ˛0 = 0
ˆ
ˆ
< P =P ˛1 = 0; ˛2 = 0
B A
(6.16)
ˆ Dis
ˆ PC = PB + PEBU
ˆ ˛3 = 0; ˛4 = 1
:̂
‍ PC  PCh
EVB = 0 ˛5 = 1 ‍
Mode 5: The EVB will get charged from the PV and grid. In this case, the output
power from EBU is zero. In this mode, the powers at the nodes are given by (6.17).
8̂
ˆ
< PA = PB = PPV ˛0 = 0; ˛1 = 0
P = P B + PgC ˛3 = 1; ˛4 = 0 (6.17)
ˆ C
:̂ P  PCh = 0 ˛5 = 1
‍ C EVB ‍
Mode 6: There is no PV power; the EVB will be charged from the EBU and the grid.
The power at the power nodes can be expressed by (6.18).
(
PC = PC Dis
g + PEBU ˛3 = 1; ˛4 = 1
Ch (6.18)
‍ PC  PEVB = 0 ˛5 = 1 ‍
Mode 7: The EBU stops discharging as it reaches the minimum SoC; the EVB
gets charged from the grid power only. The power flow at the power nodes can be
expressed by (6.19).
(
PC = PC
g ˛3 = 1; ˛4 = 0
Ch (6.19)
‍ PC  PEVB = 0 ˛5 = 1 ‍
Mode 8: Local load support is required when the grid is disconnected. The power
flow at the power nodes can be expressed by (6.20).
176 Cable based and wireless charging systems for electric vehicles

Table 6.1   Defined operation modes associated with αi

Modes ‍˛0‍ ‍˛1‍ ‍˛3‍ ‍˛4‍ ‍˛5‍ ‍˛6‍ ‍˛7‍

1 0 0 −1 0 0 0 0
2 −1 0 0 0 0 0 0
3 0 0 0 0 −1 0 0
4 0 0 0 1 −1 0 0
5 0 0 1 0 −1 0 0
6 0 0 0 1 −1 0 0
7 0 0 1 0 −1 0 0
8 0 0 0 1 0 0 1
9 0 0 0 1 0 −1 1

8̂
ˆ
ˆ PA = PPV ˛0 = 0
ˆ
ˆ
ˆ
ˆ
< PB = PA ˛1 = 0; ˛2 = 0
PC = PB + PDis ˛3 = 0; ˛4 = 1 (6.20)
ˆ
ˆ
EBU
ˆ
ˆ P = P ˛5 = 0
ˆ
ˆ D C
:̂ P = P + PDis ˛6 = 0, ˛7 = 1 ‍
‍ L D EVB

Mode 9: Grid support is required during the peak demand. The power flow at the
power8̂nodes can be expressed by (6.21).
ˆ
ˆ PA = PPV ˛0 = 0
ˆ
ˆ
ˆ PB = P A
ˆ ˛1 = 0; ˛2 = 0
<
Dis
PC = PB + PEBU ˛3 = 0; ˛4 = 1
ˆ
ˆ
ˆ
ˆ P = PC ˛5 = 0
ˆ
ˆ D
:̂ P  PE + PDis = 0 ˛6 = 1, ˛7 = 1‍
‍ D g EVB (6.21)

Figure 6.13   Mode transition of the FCI

 Reducing grid dependency of EV charging 177

The summary of the operation modes of the FCI presented above is shown in
Table 6.1.
The transition between the modes is shown in Figure 6.13 shows that αi = 0 at
all modes, this is because the FCI will have the functionality of changing the EBU
from the grid when the electricity price is low. However, in this chapter, this mode is
not considered, as the idea is to utilize the generated PV power as much as possible.

6.3.1 Power converter for FCI

The current advancement in the EV battery and semiconductor device technologies
allows the EV to travel a long distance of up to 400 km, which requires the EVB
capacity to be a few hundred kilowatts. Therefore, the power rating of the FCI needs
to be high enough to deliver the high power to an EV with less charging time. By
considering the required high power, the converter types need to be chosen carefully.
The proposed power converter of the FCI is shown in Figure 6.14.

6.3.1.1 Grid converter
The GC will play a key role in the FCI to achieve reliable operation. As discussed
in the previous analysis, the power rating of the GC should be chosen based on
(6.12) or (6.13). The three-­level neutral point clamped (3L-­NPC) topology is chosen
as it shows the benefit of having a lower current total harmonic distortion, con-
sequently facilitating the reduced volume of grid filter. The 3L-­NPC inverter has
been chosen for the GC as it has been proven for its high frequency, high voltage,
and high power in the power system applications with the power ranging up to a
few hundred kilowatts per single topology. In the past, only power IGBTs can han-
dle this level of power; however, with the fast development of the silicon carbide
metal-­oxide-­semiconductor field-­effect transistors (MOSFETs), having the voltage
and current rating about 1 kV and 100 A, respectively, the required level of power
can be achieved. Also, using this topology, the power level can be upgraded easily
by connecting more MOSFETs in parallel at each leg of the 3L-­NPC topology. The
GC based on 3L-­NPC topology is shown in Figure 6.14.

6.3.1.2 PV converter
The PV boost converter is to track the maximum power from the PV arrays. However,
for a high-­power application, the boost converter will require a big inductor, which
will result in having a high-­power loss thereby reducing the overall efficiency. To
overcome this disadvantage, an interleaved boost converter is utilized for the pro-
posed PV converter. This helps increase system efficiency by reducing the core size
of the inductor. Also, the inductor current ripple is reduced significantly. The PV
converter is shown in Figure 6.14.

6.3.1.3 EBU and EVB converter

The DAB converter is the best candidate for the EV charger. Taking advantage of its
many capabilities, the DAB converter is used as the power converter to control the
178 Cable based and wireless charging systems for electric vehicles

Figure 6.14   Power converter and control diagram of the FCI

charging and discharging of the EBU and EVB. The EBU and EVB converters are
shown in Figure 6.14.

6.3.2 Control diagram for FCI

Based on the defined operation modes, the control flowchart of the FCI is shown
in Figure 6.15. The control operation of the FCI will follow the control flowchart,
as shown in Figure 6.15. The transition between the defined modes is based on the
 Reducing grid dependency of EV charging 179

Figure 6.15   Control flowchart of the FCI

status of the grid, PV power, EBU power, and EVB power. The detailed control of
each power converter is discussed in this section.

6.3.2.1 Maximum power point tracking control

The control scheme of the PV converter is shown in Figure 6.16. There is a two-­loop
control: the outer loop for the PV voltage control Vpv and the inner loop for the boost
inductor current control IL, which is calculated according to (6.22).
‍ IL = IL1 + IL2‍ (6.22)
where IL1 and IL2 are inductor currents of the interleaved boost converter.
The two power switches will be controlled compensatively. The inductor refer-
ence current is calculated based on (6.23), where ‍K1p and ‍K1i ‍ are parameters of the
voltage controller CVpv(s).

Figure 6.16   MPPT control of the PV converter

180 Cable based and wireless charging systems for electric vehicles

Figure 6.17   The control diagram of the EVB converter

ref   K1i  ref 

‍ IL = K1p Vref
pv  Vpv + s Vpv  Vpv ‍ (6.23)

The duty cycle dpv is calculated based on (6.24), where ‍K2p ‍and ‍K2i ‍are parameters of
the current controller CIL(s). The duty cycle of the compensated switch ‍K2p ‍is calcu-
lated based on (6.25).
  K2  
ref ref
dpv = K2p IL  IL + i IL  IL (6.24)
‍ s ‍
N
dpv = 1  dpv ‍ (6.25)
‍

6.3.2.2 Electric vehicle battery control

The EVB converter control structure is depicted in Figure 6.17. Based on the selected
modes as shown in the flowchart in Figure 6.15, the EVB converter has three control
schemes. The controlled variables are calculated by (6.26)–(6.28).
There are three combinations of the operation of switches K1 and K2 correspond-
ing to the three control schemes of the EVB converter. The first scheme is the EVB
control scheme given by (6.26). This control is used during both the charging and
discharging modes of EVB with the switches K1 and K2 opened. The second scheme
given by (6.26) and (6.27) is for EVB charging at a CV, where K1 is opened and K2
is closed. The third control scheme represented by (6.26) and (6.28) involves the
control of dc-­link voltage. This control scheme is chosen in the local support mode
(Mode 8), where the EVB power is used to supply the local load. In this scheme, K1
is closed and K2 is opened.
  K3  
ref ref ref
IEVB = K3p VEVB  VEVB + i VEVB  VEVB (6.26)
‍ s ‍
  K4  
ref ref
'EVB = K4p IEVB  IEVB + i IEVB  IEVB (6.27)
‍ s ‍
  K5  
ref ref ref
IEVB = K5p Vdc  Vdc + i Vdc  Vdc (6.28)
‍ s ‍
where ‍K3p , K3i , K4p , K4i , K5i , and K5p ‍ are parameters of C V I EVB
‍ EVB , CEVB , and Cdc ‍ control-
lers, respectively.
 Reducing grid dependency of EV charging 181

Figure 6.18   The control diagram of the EBU converter

6.3.2.3 Energy buffer unit control

The control of the EBU converter is similar to the control of the EVB converter, as
shown in Figure 6.18. Based on the selected modes shown previously in the flow-
chart of Figure 6.14, the EBU converter has three control schemes. The controlled
variables are calculated by (6.29)–(6.31).
  K6  
ref ref ref
IEBU = K6p VEBU  VEBU + i VEBU  VEBU (6.29)
‍ s ‍
  K7  
ref ref
'EBU = K7p IEBU  IEBU + i IEBU  IEBU (6.30)
‍ s ‍
There are three combinations of the operation of switches K3 and K4 correspond-
ing to three control schemes of the EBU converter. The first control represented by
(6.29) is the EBU current control. This control scheme is used during both the charg-
ing and discharging modes of EBU and the switches K3 and K4 remain open during
the scheme. The second control scheme given by (6.30) and (6.31) is chosen when
the EBU is being charged in the CV mode, in this scheme, where K3 is opened and
K4 is closed. The third control scheme given by (6.29) and (6.31) is associated with
the control of dc-­link voltage. This control scheme is chosen for the local support
mode (Mode 8) where the EBU supplies power to the local load. In this scheme, K3
is closed and K4 is opened.
  K8  
ref ref ref
IEBU = K8p Vdc  Vdc + i Vdc  Vdc (6.31)
‍ s ‍
where ‍K6P , K6i , K7P , K7i , K8i , and K8p ‍ are parameters of C V I EBU
‍ EBU , CEBU , and Cdc ‍ control-
lers, respectively.

6.3.2.4 Grid converter control

The GC has two control schemes, the grid-­connected and the standalone mode as
shown in Figure 6.19. The overall control system of the GC is shown in Figure 6.19
where the space vector modulation is chosen as the modulation method due to its
lowest harmonic performance. In the grid-­connected mode, the GC will be an active
rectifier when it delivers power from the distribution grid to EVB and will be a grid
inverter when it delivers power from PV, EBU, and EVB to the distribution grid. In
this mode, a two-­loop control is deployed, the inner control loop controls the active
182 Cable based and wireless charging systems for electric vehicles

Figure 6.19   The control diagram of the GC converter

and reactive power, and the outer loop controls the dc-­link voltage Vdc. All the con-
trollers are designed in the d-­q frame as given by (6.32)–(6.34) with id, where id is
the grid current in the d-­q frame, dvdc is the control variable of the dc-­link voltage
control, and ugd and ugq are the control variables of the grid control in the d-­q frame.
 
p kivdc  ref 
dvdc = kvdc + Vdc  Vdc (6.32)
‍ s ‍
 i  
p k ref
ugd = kd + d id  id (6.33)
‍ s ‍
!
p
kiq  ref 
ugq = kq + iq  i q (6.34)
s
‍ ‍
In the standalone mode, the GC acts as an inverter that converts the dc power to ac
power to supply the local load. This is represented as mode 8, shown in the flowchart
of Figure 6.15. The detailed control of the GC in the standalone mode is shown in
Figure 6.19. The outer loop is the output voltage of the GC, vL,abc. All the controllers
are designed in the d-­q frame with iLd and iLq are the load current in the d-­q frame,
and vLd and vLq are the load voltage in the d-­q frame.
 Reducing grid dependency of EV charging 183
 
ref p ki  ref 
iLd = kLd + Ld vLd  vLd (6.35)
‍ s ‍
!
k i  
ref p Lq ref
iLq = kLq + vLq  vLq (6.36)
s
‍ ‍
 i  
p k ref
uLd = kLd + Ld iLd  iLd (6.37)
‍ s ‍
!
k i  
p Lq ref
uLq = kLq + iLq  iLq (6.38)
s
‍ ‍

6.3.3 Simulation results for FCI

To validate the performance of the proposed FCI and its control system, a detailed
model of all power converter switches is programmed in the MATLAB/Simulink
environment. Two lead-­acid battery banks rated 10 kWh are modeled in Simulink.
These battery banks are considered as the EVB and EBU, respectively. It is noted
that there is no limit in the power rating of the FCI in the simulation environment.
The main reason to choose the power rating of 10 kWh is to match with the experi-
mental system that will be implemented at the laboratory scale. The SoC of the EVB
and EBU can be calculated as given in (6.39) and (6.40):

EVB   
ˆ t
EVB 1
SoC t = SoC0 + EVB iEVB t dt (6.39)
‍ C t0 ‍
 
ˆ t
1
SoCEBU t = SoCEBU 0 + EBU
iEBU t dt (6.40)
‍ C t0 ‍
where C EVB EBU
‍ 0 and C0 ‍are the initial SoC at t = t0, C is the nominal EVB and EBU
capacities, and iEVB (t) and iEBU (t) are the EVB and EBU currents at time t. The simu-
lation results of modes 4–7 are shown in Figures 6.20 and 6.21.

Figure 6.20   Simulation results of the transition between modes 4, 5, 6, and 7

184 Cable based and wireless charging systems for electric vehicles

Figure 6.21   Simulation results of the SoC changing in modes 4, 5, 6, and 7

In mode 7, the EVB is charged from the grid only, which starts from t = 0 s
to t = 1 s. Mode 6 starts from t = 1 s to t = 2 s when the EVB is charged directly
from the EBU without the PV power while there is no requirement for grid support.
Mode 5 starts from t = 2 s to t = 3 s, which shows the charging power of the EVB
is taken from the grid and the PV while the EBU is assumed to have entered the low
SoC mode. The period from t = 3 s to t = 4 s represents mode 4, where the EVB is
charged from the EBU and PV power. The simulation results of the modes 1, 2, 3, 8,
and 9 are shown in Figures 6.22 and 6.23.
In mode 1, the EBU is charged directly from the PV power only, which starts
from t = 0 s to t = 1 s. Mode 2 starts from t = 1 s to t = 2 s when the EVB is absent
or not in the operating modes, where the PV power is injected into the grid. Mode 3
starts from t = 2 s to t = 3 s, where the charging power of the EVB is drawn from
PV only. The period from t = 3 s to t = 4 s represents mode 8, which is the load sup-
port mode where the grid is disconnected. All the distributed powers will support
the local load. The period from t = 4 s to t = 5 s represents mode 9, which is the
grid support mode, where the grid is at the peak demand and needs support. All the
distributed powers will inject the power into the grid as per request.
The simulation results show that with the proposed efficient energy manage-
ment of the distributed powers and grid power, the grid will not be tremendously

Figure 6.22   Simulation results of the transition between modes 1, 2, 3, 8, and 9

 Reducing grid dependency of EV charging 185

Figure 6.23   Simulation results of the SoC changing in modes 1, 2, 3, 8, and 9

affected by the EV fast-­charging. The distribution grid can also handle the increas-
ingly high penetration of EVs. The simulation results show that, in a future smart
grid, where the grid is to be controlled for being smarter, the proposed energy man-
agement will be more effective as the PV power generation and load demand curves
can be forecasted dynamically. This ensures that the EVB power reference can be
produced more accurately.

6.3.4 Summary designing of an FCI

An FCI using solar PV and an energy storage unit to mitigate the effects of the
high impact of EVs in all levels of charging from Level 1 to Level 3 of the EV
charging system has been proposed. The power converters of the proposed FCI
have been selected carefully, and the performance of the proposed FCI has been
investigated in detail. The control system has been designed based on the operation
modes, which are defined based on all possibilities of the exchanged power among
the EVB, EBU, PV, and the distribution grid. The status of all sources will define
each operation mode accordingly. The control of each power converter utilizes pro-
grammable switches to direct the control scheme to the corresponding mode. The
simulation results of all operation modes of the overall system, showing the transi-
tion operations of the various operation modes, demonstrate the robustness of the
proposed FCI.

6.4 Conclusions

In this chapter, an overview of Levels 1, 2, and 3 EV charging systems has been

presented. The investigation of the power converters and charging/discharging strat-
egies for the EV is shown. To reduce the effects of EV charging on the distribution
grid, an EVC-­HSP system and FCI using solar PV and an energy buffer unit have
been presented. This not only helps to mitigate the effects of the potential impact of
the integration of EVs in the distribution grid but also helps to mitigate the impact of
high penetration of PV systems. The proposed designs of the power converters and
controllers in this chapter can serve as a reference to help engineers who are working
186 Cable based and wireless charging systems for electric vehicles

in the field of renewable energy to choose suitable configurations and power con-
verters for their systems.

References

[1] Tran V.T., Islam M.R., Muttaqi K.M., Sutanto D. ‘An on-­board V2X elec-
tric vehicle charger based on amorphous alloy high-­frequency magnetic-­link
and SiC power devices’. IEEE Industry Applications Society Annual Meeting.
USA: Baltimore, MD; 2019. pp. 1–6.
[2] Lutsey N. Power play: how goverments are spurring the electric vehicle
industry, the inter. coun. clean transport [online]. Available from https://​
theicct.​org/​sites/​default/​files/​publications/​EV_​Government_​WhitePaper_​
20180514.​pdf [Accessed 20 Feb 2020].
[3] Clement-­Nyns K., Haesen E., Driesen J. ‘The impact of charging plug-­in hy-
brid electric vehicles on a residential distribution grid’. IEEE Transactions on
Power Systems. 2010, vol. 25(1), pp. 371–80.
[4] Qian K., Zhou C., Allan M., Yuan Y. ‘Modeling of load demand due to EV bat-
tery charging in distribution systems’. IEEE Transactions on Power Systems.
2011, vol. 26(2), pp. 802–10.
[5] Yilmaz M., Krein P.T. ‘Review of the impact of vehicle-­to-­grid technologies
on distribution systems and utility interfaces’. IEEE Transactions on Power
Electronics. 2013, vol. 28(12), pp. 5673–89.
[6] Dubey A., Santoso S. ‘Electric vehicle charging on residential distribution
systems: impacts and mitigations’. IEEE Access. 2015, vol. 3, pp. 1871–93.
[7] Farzin H., Moeini-­Aghtaie M., Fotuhi-­Firuzabad M. ‘Reliability studies
of distribution systems integrated with electric vehicles under battery-­
exchange mode’. IEEE Transactions on Power Delivery. 2016, vol. 31(6),
pp. 2473–82.
[8] International S. SAE’s J1772 ‘combo connector’ for ac and dc charging
advances with IEEE’s help [online]. Available from http://​articles.​sae.​org/​
10128/ [Accessed 20 Feb 2020].
[9] Kisacikoglu M.C., Kesler M., Tolbert L.M. ‘Single-­phase on-­board bidirec-
tional PEV charger for V2G reactive power operation’. IEEE Transactions on
Smart Grid. 2015, vol. 6(2), pp. 767–75.
[10] Haoyu W., Dusmez S., Khaligh A. ‘Design and analysis of a full-­bridge
LLC-­based PEV charger optimized for wide battery voltage range’. IEEE
Transactions on Vehicular Technology. 2014, vol. 63(4), pp. 1603–13.
[11] Musavi F., Edington M., Eberle W., Dunford W.G. ‘Evaluation and efficiency
comparison of front end AC-­DC plug-­in hybrid charger topologies’. IEEE
Transactions on Smart Grid. 2012, vol. 3(1), pp. 413–21.
[12] Shi C., Wang H., Dusmez S., Khaligh A. ‘A SiC-­based high-­efficiency iso-
lated onboard PEV charger with ultrawide DC-­link voltage range’. IEEE
Transactions on Industry Applications. 2017, vol. 53(1), pp. 501–11.
 Reducing grid dependency of EV charging 187

[13] Restrepo M., Morris J., Kazerani M., Canizares C. ‘Modeling and testing of
a bidirectional smart charger for distribution system EV integration’. IEEE
transactions on smart grid. 2016,1.
[14] Gautam D.S., Musavi F., Edington M., Eberle W., Dunford W.G. ‘An au-
tomotive onboard 3.3-­kW battery charger for PHEV application’. IEEE
Transactions on Vehicular Technology. 2012, vol. 61(8), pp. 3466–74.
[15] Kesler M., Kisacikoglu M.C., Tolbert L.M. ‘Vehicle-­to-­grid reactive power
operation using plug-­in electric vehicle bidirectional offboard charger’. IEEE
Transactions on Industrial Electronics. 2014, vol. 61(12), pp. 6778–84.
[16] Zhang J., Lai J.-S., Kim R.-Y., Yu W. ‘High-­power density design of a soft-­
switching high-­power bidirectional dc–dc converter’. IEEE Transactions on
Power Electronics. 2007, vol. 22(4), pp. 1145–53.
[17] Buja G., Bertoluzzo M., Fontana C. ‘Reactive power compensation ca-
pabilities of V2G-­enabled electric vehicles’. IEEE Transactions on Power
Electronics. 2017, vol. 32(12), pp. 9447–59.
[18] Tan L., Wu B., Yaramasu V., Rivera S., Guo X. ‘Effective voltage balance
control for bipolar-­DC-­bus-­fed EV charging station with three-­level DC–DC
fast charger’. IEEE Transactions on Industrial Electronics. 2016, vol. 63(7),
pp. 4031–41.
[19] Rivera S., Wu B. ‘Electric vehicle charging station with an energy storage
stage for split-­DC bus voltage balancing’. IEEE Transactions on Power
Electronics. 2017, vol. 32(3), pp. 2376–86.
[20] Cao Y., Tang S., Li C. ‘An optimized EV charging model considering TOU
price and SOC curve’. IEEE Transactions on Smart Grid. 2012, vol. 3(1), pp.
388–93.
[21] Richardson P., Flynn D., Keane A. ‘Optimal charging of electric vehicles in
low-­voltage distribution systems’. IEEE Transactions on Power Systems.
2012, vol. 27(1), pp. 268–79.
[22] Deilami S., Masoum A.S., Moses P.S., Masoum M.A.S. ‘Real-­time coordina-
tion of plug-­in electric vehicle charging in smart grids to minimize power
losses and improve voltage profile’. IEEE Transactions on Smart Grid. 2011,
vol. 2(3), pp. 456–67.
[23] Chukwu U.C., Mahajan S.M. ‘V2G parking lot with PV rooftop for capac-
ity enhancement of a distribution system’. IEEE Transactions on Sustainable
Energy. 2014, vol. 5(1), pp. 119–27.
[24] Alam M.J.E., Muttaqi K.M., Sutanto D. ‘Effective utilization of available
PEV battery capacity for mitigation of solar PV impact and grid support with
integrated V2G functionality’. IEEE Transactions on Smart Grid. 2016, vol.
7(3), pp. 1562–71.
[25] Foster J.M., Trevino G., Kuss M., Caramanis M.C. ‘Plug-­in electric vehi-
cle and voltage support for distributed solar: theory and application’. IEEE
Systems Journal. 2013, vol. 7(4), pp. 881–8.
[26] Badawy M.O., Sozer Y. ‘Power flow management of a grid tied PV-­battery
system for electric vehicles charging’. IEEE Transactions on Industry
Applications. 2017, vol. 53(2), pp. 1347–57.
188 Cable based and wireless charging systems for electric vehicles

[27] Islam M.R., Rahman M.F., Xu W. Advances in solar photovoltaic power

plants: Green Energy and Technology Series. Germany: Springer-­Verlag
GmbH, Heidelberg; 2016.
[28] Abdelrazek S.A., Kamalasadan S. ‘Integrated pv capacity firming and energy
time shift battery energy storage management using energy-­oriented optimi-
zation’. IEEE Transactions on Industry Applications. 2016, vol. 52(3), pp.
2607–17.
[29] Tran V.T., Islam M.R., Muttaqi K.M., Sutanto D. ‘An efficient energy man-
agement approach for a solar-­powered EV battery charging facility to support
distribution grids’. IEEE Transactions on Industry Applications. 2019, vol.
55(6), pp. 6517–26.