Energy 132 (2017) 1e11

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Novel grid-connected solar/wind powered electric vehicle charging

station with vehicle-to-grid technology
Hassan Fathabadi
School of Electrical and Computer Engineering, National Technical University of Athens (NTUA), Athens, Greece

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, a novel grid-connected solar/wind powered electric vehicle charging station with vehicle-
Received 22 September 2016 to-grid technology is designed and constructed. The solar/wind powered electric vehicle charging station
Received in revised form consists of a photovoltaic array, a wind energy conversion system, two unidirectional (Direct Current)/
10 April 2017
(Direct Current) converters connected to the photovoltaic array and wind energy conversion system, a
Accepted 30 April 2017
unified maximum power point tracking controller, 15 bidirectional (Direct Current)/(Direct Current)
Available online 2 May 2017
converters associated with the 15 charging stations provided for charging electric vehicles, and a three-
phase bidirectional (Direct Current)/(Alternating Current) inverter connected to the grid. The contribu-
Keywords:
Electric vehicle
tion of this work is that the grid-connected solar/wind powered electric vehicle charging station pre-
Charging station sented in this work is the only large-scale constructed charging station reported in the literature that
Solar/wind combines two renewable resources (solar and wind) together to produce electric power to charge electric
Maximum power point tracking vehicles, and moreover, it maximally converts solar and wind energy into electric energy because it uses
a novel fast and highly accurate unified maximum power point tracking technique that concurrently
tracks the maximum power points of both photovoltaic array and wind energy conversion system. It is
demonstrated that the constructed charging station not only provides electric energy to charge electric
vehicles but also balances load demand in the grid connected to it.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
The solar/wind powered electric vehicle (EV) charging station
designed and implemented in this study consists of a photovoltaic
(PV) system and a wind energy conversion system (WECS) both
controlled by a unified maximum power point tracking (MPPT)
controller, so a concise survey is performed about different MPPT
methods applicable to PV systems and WECSs. In a PV system, a
MPPT controller tracks the maximum power point (MPP) of the PV
panel/array, and so enhances the energy efficiency [1]. Different
MPPT methods have been reported in the literature [2]. Open-
circuit voltage (OCV) method first measures the open-circuit
voltage of the PV module disconnected from the system, and
then estimates the MPP voltage [3]. Temperature method estimates
the open-circuit voltage using the PV module temperature [4].
Short-circuit current (SCC) method is very similar to the OVC
method [5] that uses the short-circuit current of the PV module to
estimate its MPP [6]. Fuzzy methods first convert the voltage and
E-mail address: h4477@hotmail.com.
current of the PV module into fuzzy variables [7], and then fuzzy
roles are used to track the MPP [8]. A modified fuzzy algorithm is
adaptive fuzzy MPPT method [9]. Artificial neural network (ANN)
based MPPT methods use the neural networks trained by the real
data presenting the MPP voltage and current under different con-
ditions. For instance, two feed-forward neural networks having two
hidden layers were used for MPPT process in Ref. [10]. Perturb and
observe (P&O) technique uses the perturbations of the output
voltage and current of the PV panel/array to find its MPP [11].
Basically, three versions of this technique are available: P&O fixed
step-size, P&O variable step-size and three-point weighted [12]. An
improved version of the P&O method with variable step-size is also
available in which a dynamic perturbation step-size is utilized to
mitigate oscillations around the MPP, while boundary conditions
are introduced to prevent it from diverging away from the MPP
[13]. Two modified hybrid methods called “particle swarm opti-
mization adaptive neuro-fuzzy inference system (PSO-ANFIS)” and
“P&O-ANFIS” were introduced and compared in Ref. [14]. The slope
of the P-V curve is used to find the MPP of the curve in incremental
conductance (IC) method [15]. Extremum seeking control (ESC)
method uses a nonlinear feedback system to estimate the MPP [16].

http://dx.doi.org/10.1016/j.energy.2017.04.161
0360-5442/© 2017 Elsevier Ltd. All rights reserved.
2 H. Fathabadi / Energy 132 (2017) 1e11
Ripple-based ESC method is a modified version of the ESC tech-
nique that is applicable only to grid-connected PV systems [17].
Power management (PM) MPPT method that is more applicable to
shaded situations [18] and scanning method which finds the MPP
by scanning and comparing output powers [19] are the two other
methods. A sensor based MPPT method first measures solar irra-
diance using sensors, and then estimates the MPP [20]. Hybrid
prediction-P&O method combines the direct-prediction and P&O
techniques to find the MPP [21]. To increase the MPPT efficiency
and tracking speed, the cuckoo search (CS) algorithm can be used to
find the MPP [22]. Field program gate array (FPGA) based MPPT
method finds the MPP by comparing the instant output power with
the calculated maximum power [23]. Auto-scaling variable step-
size IC MPPT method uses a judgment criterion and auto-scaling
variable step-size to enable the PV system to achieve a fast dy-
namic response and stable output power [24]. Two researches
showed that genetic algorithm [25] and predictive model [26] can
be also used in a repetitive process to find the MPP. In a PV system,
in addition to maximum power extraction, the MPPT controller can
be also used to control other devices [27] such as a solar tracker [28]
in which the amount of the maximum power extracted from the PV
panel is used to find the sun direction [29].
A WECS applicable to industrial applications consists of a wind
turbine, a three-phase electric generator, a rectifier, and a MPPT
controller [30]. A wind turbine converts wind energy into me-
chanical energy [31], so that, the turbine output mechanical power
gets to maximum at an optimum turbine speed [32]. A MPPT
controller regulates the speed of a turbine to its optimum speed
[33]. There are different MPPT methods applicable to WECSs [34],
but these methods such as ESC and optimal torque control (OTC) all
can be classified into the three main categories known as tip speed
ratio (TSR) control based methods that regulate the TSR of a wind
turbine [35], power signal feedback (PSF) control based methods
that regulate the output power of a wind turbine to a specific
reference value [36], and hill-climb search (HCS) technique also
known as “P&O method” that tracks the MPP of a wind turbine by
producing a set of perturbations in the turbine speed [37]. In the
TSR control based methods, the wind and turbine speeds both
should be first measured by two speed sensors (anemometer and
tachometer), and then, the TSR of the turbine is regulated to the
optimum TSR which is a specific parameter of the turbine [38]. The
PSF control based methods use a mechanical power equation or
power curve indicating the turbine speed and the related
maximum power [39]. This power curve is found by performing an
experiment or simulation. When the turbine operates, the
maximum power specified by the curve is used as a reference value
to regulate the turbine output instant power to the reference value
by varying the turbine speed. Similarly, the OTC method uses a
mechanical torque equation or torque curve to obtain a reference
value for the optimum torque associated with the turbine instant
speed [40]. The TSR and PSF control based methods need speed
sensors to accurately track the MPP, otherwise they are not accurate
enough [41]. The HCS method was firstly applied to PV systems
[42], it produces a perturbation in the turbine speed, and then
compares the current and previous amounts of the turbine output
power to determine the next perturbation [43]. The convergency of
the method significantly depends on turbine’s current operating
point and the perturbation type, so that, the method may diverge or
even converge to an incorrect point [44]. Some hybrid MPPT
methods are also available that are applicable to hybrid systems
such as PV/fuel cell hybrid systems [45]. In fact, two or more of the
MPPT methods explained in detail in this section have been com-
bined to each other to form a hybrid MPPT method [46].
Different types of EV charging stations have been reported in the
literature [47], most of them only use solar energy such as the EV
charging station design reported in Ref. [48], and all are simulation
based works (models) [49]. Some EV charging stations reported in
the literature also use wind energy solely or combined with other
power sources, they are in fact hybrid types [50], but similarly, they
are only some ideal simulated models [51]. Moreover, there is not
any new MPPT consideration in the models presented in the pre-
vious works [52]. In this study, a grid-connected solar/wind pow-
ered EV charging station with vehicle-to-grid (V2G) technology is
designed and constructed. It is the only large-scale constructed EV
charging station reported in the literature that uses solar and wind
energy to produce electric power to charge EVs. Moreover, the
charging station maximally converts solar and wind energy into
electric energy because it uses a novel fast and highly accurate
unified MPPT technique that concurrently tracks the maximum
power points of both PV system and WECS. The method only uses
the output voltages and currents of the PV array and WECS, i.e., it
neither needs any sensor such as anemometer nor has the draw-
backs of the other MPPT methods explained in detail. It is also
demonstrated that the EV charging station not only provides
electric energy to charge EVs but also balances load demand in the
grid [53]. The rest of this paper is organized as follows. The design
and implementation of the solar/wind powered EV charging station
is explained in detail in Section 2. Constructed system and exper-
imental verifications are presented in Section 3, and Section 4
concludes the paper.
2. Design and implementation of the solar/wind powered EV
charging station
The grid-connected solar/wind powered EV charging station has
been designed, constructed and located where on-shore wind
blows with an average speed of 41.6 km/h almost during the whole
of the year. The weather statistical data extracted from weather
forecast reports during the last 5 years (2011e2016) also demon-
strate that there are averagely 273 sunny days during the year. The
first step was to estimate the daily electric power that should be
produced or provided by the EV charging station. The average
number of the plug-in hybrid electric vehicles (PHEVs) which
should be charged and the daily time of charging each PHEV have
been estimated based on the results of a statistical research per-
formed around the charging station location. First, the names and
addresses of the owners of the PHEVs living in a radius of 15 km
around the charging station were obtained from the two only sale
agents. Then, the number of the PHEVs and the daily time that each
owner usually intends to charge his/her PHEV were extracted from
the questionnaires distributed between the owners. The outcome
of the research showing the number of the PHEVs that should be
charged hour by hour during a day is shown in Fig. 1(a). To estimate
the charging profile, loading duration of each PHEV is considered
one hour, and 1.2 kWh is chosen as a basic number representing the
basic amount of the electric energy needed to be supplied to each
PHEV. Thus, the electric power needed to charge the PHEVs hour by
hour (charging profile) can be estimated by multiplying the PHEVs
number by 1.2 kW as shown in Fig. 1(b). In the estimation, a PHEV
with the battery capacity more than the basic number has been
considered as several PHEVs. For instance, by assuming SOC ¼ 0.2
(state of charge) as the battery level at which a PHEV starts to be
charged, a 5.2 kWh PHEV has been approximately considered as 3
PHEVs. In Fig. 2, the charging profile has been divided into two
parts, the green and blue parts. The green portion of the charging
profile is produced using solar energy in daylight, while the blue
portion is provided by converting wind energy into electric power.
The configuration of the designed solar/wind powered EV charging
station is shown in Fig. 3. The solar/wind powered station consists
of a PV array, a WECS, two unidirectional DC/DC converters
 H. Fathabadi / Energy 132 (2017) 1e11
Fig. 1. (a) PHEVs number per hour that should be charged during a day. (b) Estimated
daily charging profile.
Fig. 2. Charging profile: Green portion is produced using solar energy, and the blue
portion is produced using wind energy. (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of this article.)
dedicated to the PV array and WECS, a unified MPPT controller, 15
bidirectional DC/DC converters associated with the 15 charging
stations provided for charging EVs, and a bidirectional DC/AC
inverter connected to the grid. It can be seen that all the 15 EV
charging stations have the capability of vehicle-to-grid (V2G)
connection that means each EV not only can be charged but also
may be even discharged to help the grid to provide enough electric
energy for consumers when load demand peaks in the grid.
Considering the green portion of the charging profile shown in
3
Fig. 3. Configuration of the proposed solar/wind powered EV charging station.
Fig. 2 and environmental conditions such as cloudy sky, a 12 kW PV
array has been chosen for the PV system. Similarly, by noting the
blue portion of the charging profile and the availability of wind
blow with the average speed of 41.6 km/h, a 10 kW vertical wind
turbine with the rated wind speed of 12 m/s has been chosen for
the WECS. The PV array and WECS both have been connected to the
DC bus through the two identical unidirectional DC/DC converters,
and the MPPT controller concurrently tracks the maximum power
points of the PV array and WECS. To track the two maximum power
points, the DC-link voltage is used as a reference voltage by the
MPPT controller. The circuit of the two identical unidirectional DC/
DC boost converters connected to the PV array and WECS is shown
in Fig. 4 [54]. It is a high efficient DC/DC converter including only
one N-MOSFET switch S1 . The switch S1 operates with a constant
switching frequency of fs and a variable duty cycle of DS ¼ ton =Ts ,
where Ts ¼ 1=fs is switching period, and ton is the switch S1 on-
time. The gain of this DC/DC converter is obtained as [54]:
Vdc n
¼ (1)
Vin 1  DS
where n and Vin are respectively the transformer ratio and the
converter input voltage, and Vdc is the converter output voltage as
shown in Fig. 4. The implementation of the solar/wind powered EV
charging station is shown in detail in Fig. 5. The outputs of the two
unidirectional DC/DC converters dedicated to the PV array and
WECS are connected to the DC bus, so in fact, Vdc is the DC-link
voltage provided by the two unidirectional DC/DC converters. The
different parts of the solar/wind powered EV charging station are
explained and analyzed in detail in separate sub-sections as
follows.
2.1. PV system and MPPT process
As shown in Fig. 5, the PV system consists of the PV array and its
associated unidirectional DC/DC converter. The output power ðPpv Þ
of the PV array shown in Fig. 5 is obtained as:
Ppv ¼ Vpv Ipv (2)
where Vpv and Ipv are the PV voltage and current, respectively. The
P-V characteristic of the PV array used in this study is shown in
Fig. 6 where the P-V curve has been divided into the three parts: the
MPP, the left side of the MPP, and the right side of the MPP. At the
MPP, the PV output power derivative is zero, so:
4 H. Fathabadi / Energy 132 (2017) 1e11

Fig. 4. Unidirectional DC/DC boost converter connected to the PV array and WECS.

Fig. 5. Implementation of the solar/wind powered EV charging station.

dPpv
>0 (4)
dVpv
Similarly, at the right side, it is negative, so:

dPpv
<0 (5)
dVpv
The PV power derivative is approximated by defining power
slope apv as:

dPpv DPpv Ppv ðkÞ  Ppv ðk  1Þ

z ¼ ¼ apv (6)
dVpv DVpv Vpv ðkÞ  Vpv ðk  1Þ

where Vpv ðkÞ and Ipv ðkÞ are respectively the kth samples of the PV
voltage and current measured by the MPPT controller, and Ppv ðkÞ is
the associated PV output power which is calculated as:

Ppv ðkÞ ¼ Vpv ðkÞIpv ðkÞ (7)

Fig. 6. P-V characteristic of the PV array used in this study under nominal condition
(solar irradiance G ¼ 1000 Wm2 , cell temperature T ¼ 25  C). The relationship between the DC-link voltage and the PV voltage
can be expressed using Eq. (1) as:


dPpv 1  Dpv Vdc
¼0 (3) Vpv ¼ (8)
dVpv n

At the left side of the MPP, the derivative is positive, so: where Dpv is the duty cycle of the DC/DC unidirectional DC/DC
 H. Fathabadi / Energy 132 (2017) 1e11 5

converter connected to the PV array. Since the DC-link voltage is significantly decreases the amount of time it takes to move the
constant, the PV voltage Vpv can be regulated to the PV voltage at operating point to around the MPP, and so minimizes the tracking
the MPP ðVpvmpp Þ by varying the duty cycle Dpv , so that, Vpv is convergence time. By reduction   in the distance between the
increased by deceasing Dpv , and similarly, Vpv is decreased by operating point and the MPP, apv  becomes lower and lower, and so
increasing Dpv . Thus, MPPT can be performed in the PV system by DDpv is chosen smaller and smaller to maximize the MPPT effi-
only varying the duty cycle Dpv . The flowchart indicating the MPPT ciency (accuracy of the MPPT process).
process in the PV system performed by the MPPT controller is Tracking the MPP of a PV array becomes more difficult when
shown in Fig. 7. The PV voltage ðVpv ðkÞÞ and current ðIpv ðkÞÞ are there is not a homogeneous solar irradiance on the PV panel’s
continually measured, and the PV output power ðPpv ðkÞÞ is surface because of partial shading, i.e., one or more PV modules are
computed using Eq. (7) by the MPPT controller, then the power partially shaded, and so receive less solar irradiance compared to
slope apv is calculated using Eq. (6). If apv ¼ 0 (at the MPP), there is the others. Under partial shading conditions, one global MPP
no need to change Dpv . If apv > 0 (left side of the MPP), Vpv is (GMPP) and one or more local MPPs generally occur on the P-V
continually increased by decreasing Dpv , i.e., characteristic of the PV panel because of the operation of the bypass
Dpv ðk þ 1Þ ¼ Dpv ðkÞ  DDpv , where DDpv is a positive amount that diodes. The presence of the multiple peaks causes an essential
will be explained next. If apv < 0 (right side of the MPP), Vpv is problem, so tracking the GMPP becomes more complicated. The
continually decreased by increasing Dpv , i.e., unified MPPT technique presented in this work tracks the GMPP,
Dpv ðk þ 1Þ ¼ Dpv ðkÞ þ DDpv , until the PV operating point reaches the not the LMPPs, under partial shading conditions because it uses the
MPP. When DVpv ¼ 0, the MPPT controller waits for the next sample following strategies:
of the PV module voltage as shown in the flowchart. As mentioned,
the increase/decrease in Dpv is performed by adding/subtracting the A) At the beginning (t ¼ 0), when the MPPT process starts in the
positive amount DDpv to/from Dpv . More than 50 experiments were PV system implemented in the EV charging station, the
performed for the constructed PV system under different condi- following steps are performed:
tions (different solar irradiances and temperatures) that demon- Step 1: Tracking the MPP is first performed with a spec-
strated choosing DDpv according to the following equation provides ified initial value of the converter’s duty cycle which is
not only a minimum tracking convergence time but also a 0.07, i.e. Dpv ¼ 0:07.
maximum MPPT efficiency: Step 2: After finding the MPP, tracking the MPP is
repeated 14 times with 14 distinct initial values of the
8   duty cycle that are 0.16, 0.24, 0.33, 0.39, 0.42, 0.47, 0.53,
> 0:100 when apv   40 WV1
>   0.59, 0.66, 0.71, 0.77, 0.82, 0.88 and 0.94. Thus, 15 MPPs
>
>
>
< 0:050 when 20 WV1  apv  < 40 WV1 are totally found, and the voltage associated with each
 
DDpv ¼ 0:010 when 10 WV1  apv  < 20 WV1 (9) MPP is stored in the RAM of the microcontroller.
>
>  
>
> 0:005 when 5 WV  apv  < 10 WV1
1 Step 3: After that, the output powers associated with the
>
:   15 MPPs are compared to each other to distinguish the
0:001 when 0 WV1  apv  < 5 WV1
GMPP from the local MPPs, and the voltage associated
Eq. (9) demonstrates that with the GMPP is extracted from the RAM, so the MPPT
 when the MPP is far from the PV ar-
ray’s operating point, apv  is high, so to decrease the tracking controller tracks the GMPP, not the local MPPs.
convergence time, DDpv is selected larger. This selection B) In addition to the beginning of the MPPT process (t ¼ 0), the
above-mentioned three steps are also performed by the
proposed MPPT technique whenever a sudden variation
more than 100 W occurs in the output power of the PV array.
This sudden variation plays the role of an alarm notifying the
occurrence of partial shading.

2.2. Wind energy conversion system and MPPT process

As shown in Fig. 5, the WECS used in the solar/wind powered EV

charging station consists of a wind turbine, a permanent magnet
synchronous generator (PMSG) [55], a three-phase diode bridge
rectifier, and the dedicated unidirectional DC/DC converter. For the
wind turbine used in the WECS, the tip speed ratio (TSR) is obtained
as [56,57]:

R um
l¼ (10)
Vw

where Vw is the wind speed, R is the turbine radius, and um is the

turbine angular speed. Since the pitch angle of the wind turbine
used in this study is zero, the output mechanical power of the
turbine ðPm Þ is found as [57]:

Pm ¼ Cp ðlÞPw (11)

Fig. 7. Flowchart of the MPPT process in the PV system and WECS performed by the where Pw is the wind power that rotates the turbine and Cp is the
MPPT controller. power coefficient. The output mechanical power of the 10 kW
6 H. Fathabadi / Energy 132 (2017) 1e11

vertical wind turbine used in the WECS versus the turbine angular
speed is shown in Fig. 8. The following first-order differential
equation should be satisfied by the wind turbine [56]:

dum
Tm  TL  f um ¼ J (12)
dt

where f is the viscous friction coefficient, J is the total inertia of the

wind turbine and generator’s shaft, Tm ¼ Pm =um is the turbine
mechanical torque, and TL is the electromagnetic torque which is
induced by the generator. The input power ðPwe Þ of the unidirec-
tional DC/DC converter connected to the WECS is expressed as [56]:

 1

Pwe ¼ Vwe Iwe ¼ Pw Ctgr l; um ; id ; iq ¼ rAVw
3
Ctgr l; um ; id ; iq
2
(13)

where Vwe and Iwe are respectively the input voltage and current of
the unidirectional DC/DC converter as shown in Fig. 5, id and iq are
the output currents of the generator in the ðdqÞ reference frame,
Ctgr ðl; um ; id ; iq Þ is the turbine-generator-rectifier power coefficient Fig. 9. Pwe  l characteristic of the implemented WECS at the rated wind speed (12 m/
s).
firstly introduced in Ref. [56], r is the density of the air around the
turbine, and A is the turbine’s cross-sectional area. The Pwe  l
characteristic of the implemented WECS at the rated wind speed
(12 m/s) is shown in Fig. 9. When the converter input voltage is
increased, the generator output current decreases, and hence the dPwe
>0 (15)
electromagnetic torque induced by the generator ðTL Þ decreases, dVwe
and this increases the turbine speed according to Eq. (12). Similarly, At the right side of the MPP:
when the converter input voltage is decreased, this causes a
reduction in the turbine speed. Thus, as shown in Fig. 9, the
dPwe
following conditions should be satisfied on the different parts of <0 (16)
the Pwe  l curve: dVwe
At the MPP ðlopt ; Pwemax Þ: Similar to the PV system, the power slope awe is defined as:

dPwe DPwe Pwe ðkÞ  Pwe ðk  1Þ

dPwe
¼0 z ¼ ¼ awe (17)
dVwe
(14) dVwe DVwe Vwe ðkÞ  Vwe ðk  1Þ

At the left side of the MPP: where Vwe ðkÞ and Iwe ðkÞ are respectively the kth samples of the
converter input voltage and current, and Pwe ðkÞ is the converter
input power that is computed by the MPPT controller as:

Pwe ðkÞ ¼ Vwe ðkÞIwe ðkÞ (18)

The relationship between the constant DC-link voltage and the
output voltage of the WECS ðVwe Þ shown in Fig. 5 is obtained using
Eq. (1) as:

ð1  Dwe ÞVdc
Vwe ¼ (19)
n

where Dwe is the duty cycle of the unidirectional DC/DC converter

Fig. 8. Turbine output mechanical power vs. the turbine angular speed.
dedicated to the WECS. Eq. (19) shows that the MPPT technique
used for the PV system, the flowchart of which is shown in Fig. 7,
can be again utilized to track the MPP of the WECS. Thus, the MPPT
controller similarly varies the duty cycle Dwe to regulate the WECS
output voltage to its voltage at the MPP ðVwecmpp Þ. The increase/
decrease in Dwe is performed by adding/subtracting a positive
amount DDwe to/from Dwe . Similar to the PV system, more than 50
experiments were performed for the constructed WECS under
different conditions (different wind speeds and temperatures) that
demonstrated choosing DDwe according to the following equation
provides a minimum tracking convergence time along with a
maximum MPPT efficiency:
 H. Fathabadi / Energy 132 (2017) 1e11 7

8
> 0:100 whenjawe j  20 WV1
>
>
>
< 0:050 when 10 WV1  jawe j < 20 WV1
DDwe ¼ 0:010 when 5 WV1  jawe j < 10 WV1 (20)
>
>
>
> when 2 WV1  jawe j < 5 WV1
: 0:005
0:001 when 0 WV1  jawe j < 2 WV1

3. Constructed system and experimental results

As shown in Fig. 5, there are 15 stations for charging EVs. To

provide the capability of V2G connection for all the 15 stations, a
bidirectional DC/DC converter has been connected between each
charging station and the DC bus the electric circuit of which is
shown in Fig. 10, so 15 identical bidirectional DC/DC converters
have been utilized. A separate microcontroller controls in detail
the operation of the 15 bidirectional DC/DC buck-boost converters.
The microcontroller produces a set of control signals to set each
converter in charging or discharging mode by respectively turning
on or turning off the IGBT SB. Then, the microcontroller regulates
the charging/discharging current of each converter to the required
amount by varying the duty cycle Dchar /Ddisc of the control signals
supplied to the converter’s terminal denoted as “CHARGE-1”/
“DISCHARGE -1” in Fig. 10. The DC bus of the system is also con-
nected to the grid through a conventional three-phase bidirec-
tional PWM six-switch DC/AC inverter. A protection circuit which
is needed to protect the EV charging station against the different
faults occurring in the grid such as ground, phase to phase and
short-circuit faults has been also installed in the grid side of the
inverter [58,59]. Based on the implemented solar/wind powered
EV charging station shown in detail in Fig. 5, the EV charging
station has been built. The electric circuit of the constructed solar/
wind powered EV charging station is shown in Fig. 11. A micro-
controller MC68HC11E9 has been used to implement the unified
MPPT controller. The flowchart shown in Fig. 7 has been utilized to
program the microcontroller to find the maximum power points
of both PV system and WECS. In each system, the converter input
current is continually measured by an INA 168, and is supplied to
the microcontroller through an analog/digital (A/D) pin. The con-
verter input voltage is first scaled by a potentiometer (RV1 and Fig. 11. Electric circuit of the constructed solar/wind powered EV charging station.

Fig. 10. Bidirectional DC/DC converter connected between each charging station and the DC bus.
8 H. Fathabadi / Energy 132 (2017) 1e11

RV2), and then, is supplied to an A/D pin of the microcontroller.

The microcontroller continually samples the input currents and AveragepowerextractedbytheMPPTcontroller
voltages of the two converters with a sampling period of 0.5 ms, MPPTefficiency¼
Maximumavailablepower
and determines the duty ratios Dpv and Dwe according to the
100
flowchart of the MPPT process shown in Fig. 7. Then, the specified
duty ratios Dpv and Dwe are supplied to the N-MOSFET switches Spv 199:28W
¼ 100¼99:60%
and Swe as two periodic switching pulses with the duty cycles of 200:08W
respectively Dpv and Dwe through the two current transformers (21)
Trans-PV and Trans-WE to regulate the input powers of the two
On the other hand, the waveform shows that the tracking
converters (Ppv and Pwe ) to their maximum values (Ppvmax and
convergence time is only 12 ms, and thus, the MPPT technique
Pwemax ) by varying the converters input voltages (Vpv and Vwe ).
implemented in PV system is a very fast and high-accurate tech-
The parameters and specifications of the components used in the
nique. Similarly, to evaluate the performance of the MPPT tech-
constructed solar/wind powered EV charging station are summa-
nique to track the MPP of the WECS, at first a small 600 W vertical
rized in Table 1. The PV array of the constructed EV charging
wind turbine (model: WKV-600) was coupled to the system. The
station comprises 10 panels, each panel itself includes 6 PV
wind turbine was located in a wind tunnel, and the wind speed was
modules KC200GT, so the PV array totally consists of 60 PV
varied according to the pattern shown in Fig. 14. The waveform of
modules KC200GT. The 6 PV modules located on each panel are
the electric power extracted from the WECS by the MPPT controller
series-connected, and the 10 panels have been connected to each
is shown in Fig. 15 that again verifies the technique is highly
other in parallel to provide appropriate voltage and current levels
accurate.
[60]. The specifications of a PV module KC200GT has been re-
After installing the EV charging station, different parameters
ported in Table 1. At first, to evaluate the performance of the MPPT
were measured daily. The electric power produced by the charging
technique implemented in the PV system and WECS, a small-scale
station during a sunny day in summer, when the average wind
experimental set-up was installed by connecting only one PV
speed was about 12 m/s, is shown as curve (a) in Fig. 16, and the
module KC200GT and a small 600 W vertical wind turbine (model:
electric power consumed to charge EVs is shown as curve (b) in
WKV-600) to the system, the photography of the experimental
Fig. 16. When the electric power produced by the EV charging
set-up is shown in Fig. 12. The experimental waveform of the
station is more than the electric power consumed by the EVs, the
electric power extracted by the MPPT controller in response to a
additional power is supplied to the grid as shown in Fig. 16 (curve
positive solar irradiance step from 0 Wm2 to 1000 Wm2 under
(c)). It is clear that, when the electric power consumed to charge the
nominal condition (solar irradiance G ¼ 1000 Wm2 , cell tem-
EVs is more than the power production of the EV charging station,
perature T ¼ 25 + C) is shown Fig. 13. The electric power extracted
the supplemental electric power should be provided by the grid. In
from the PV module was also measured using a DC wattmeter
other experiment, the above-mentioned powers were measured
model DW-6064 with 0.01 W resolution. On the one hand, the
during a cloudy day in winter, when the average wind speed was
average electric power extracted by the MPPT controller is
about 11 m/s. The electric power produced by the EV charging
199.28 W, and as reported in Table 1, the maximum available
station is shown as curve (a) in Fig. 17, and the electric power
power of the PV module under nominal condition is 200.08 W, so
consumed by the EVs is shown as curve (b) in Fig. 17. Curve (c)
the MPPT efficiency of the MPPT technique under nominal con-
shows the electric power supplied to the grid, and the supple-
dition is calculated as:
mental electric power provided by the grid is shown as curve (d).
The average annual load demand in the local urban grid is also

Table 1
Specifications of the components used in the constructed solar/wind powered EV charging station.

Permanent magnet synchronous generator (PMSG) Two DC/DC boost converters connected to the PV array and WECS

Model NH-10K-150 Cin ðmFÞ-Aluminum electrolytic capacitor/400 V 470

Winding type Three-phase, star connection Converter switching frequency: fs (kHz) 25
Rated power (kW) 10 Cout ðmFÞ-Aluminum electrolytic capacitor/600 V 220
Rated phase voltage (V) 110 DC-link voltage Vdc (V) 540
Rated speed (rpm) 150 C2 (m F)-Premium metallized polypropylene capacitor/600 V 22
Nominal phase current (A) 30 C1 ðnFÞ-Parasitic capacitance of IRFPS40N60K WECS: 19.2 PV: 24
Start-up torque (Nm) <13 Type of transformer T Pulse
Magnetic material NdFeB n ¼ N2 =N1 9=6
Insulation F MOSFET switch S1 IRFPS40N60K WECS:
16 PV: 20
Weight (kg) 423 Equivalent series resistance of Cout ðmUÞ 478
Wind turbine Diodes: D1 eD2 15ETH06S WECS:2 PV:2
Model WKV-10000 PV array: 60 PV modules KC200GT
Type Vertical PV module KC200GT
Rated power (kW) 10 Current at MPP Ipvmpp (A) 7.61
Size: height/diameter (m) 4.2/5.7 Voltage at MPP Vpvmpp (V) 26.3
Blades material Al. alloy Output power at MPP Ppvmpp (W) 200.1430
Start-up wind speed (m/s) 1.5 Short-circuit current ISC (A) 8.21
Minimum power generation 2.5 Open-circuit voltage VOC (V) 32.9
wind speed (m/s)
Rated wind speed (m/s) 12 Three-phase rectifier
Cut-in wind speed (m/s) 4 Diodes D3-D8 40HF60
Cut out wind speed (m/s) 15 Rectified current (A) 40
Survival wind speed (m/s) 60 Peak repetitive reverse voltage (V) 600
 H. Fathabadi / Energy 132 (2017) 1e11 9

Fig. 13. Power extracted from one PV module KC200GT by the MPPT controller in
response to a positive solar irradiance step from 0 Wm2 e1000 Wm2 .

Fig. 14. Pattern of the wind speed variation.

Fig. 12. Small-scale experimental set-up: (a) Control unit constructed to be utilized in
the small-scale experimental set-up. (b) PV module connected to the control unit. (c)
Small 600 W vertical wind turbine (model: WKV-600) placed in the wind tunnel and
coupled to the system.

shown in Fig. 18. The following points are deduced by considering

the experimental results shown in Figs. 16e18:

 During a sunny day, the EV charging station produces enough

electric power, so that, the extra power produced by the station
is supplied to the local urban grid. Thus, it helps the grid to
provide enough electric energy in response to the load demand
in the grid, in particular, around 20e23 o’clock when the load
demand peaks in the local urban grid as shown in Fig. 18. It is
reminded that in an urban power distribution network, the peak
of the load demand generally occurs around 20e23 o’clock
when a huge amount of lighting loads is added to the grid [61].
 As shown in Fig. 17, during a cloudy day, the EV charging station
does not produce enough electric power to charge EVs, so the Fig. 15. Experimental waveform of the electric power extracted from the WECS by the
grid should provide the supplementary electric power needed. MPPT controller.
10 H. Fathabadi / Energy 132 (2017) 1e11
Fig. 16. Experimental results: Different powers measured hour by hour during a sunny
day in summer.
Fig. 17. Experimental results: Different powers measured hour by hour during a cloudy
day in winter.
But it is very important to note that the EV charging station uses
the supplemental electric power produced by the grid around
7e17 o’clock when the load demand in the grid is low. It is
worthwhile to note that similar to a sunny day, the extra power
produced by the charging station is supplied to the grid around
20e23 o’clock when the load demand peaks in the grid. Thus,
the EV charging station balances the load demand in the grid
during a sunny or cloudy day, and this is the other advantage of
installing the charging station.
Fig. 18. Average annual load demand in the local grid.
The total money invested to construct the EV charging station is
about $52500, and the average daily production of the EV charging
station is about 350 kWh. The price of the electric energy sold is
averagely $42 per 100 kWh, so the annual income is about $53500.
This means that the invested money has returned after only one
year.
4. Conclusion
In this study, a perfect grid-connected solar/wind powered EV
charging station with V2G technology was implemented. It opti-
mally uses solar and wind energies to produce electric energy to
charge EVs. A novel fast and highly accurate unified MPPT tech-
nique has been utilized to track the maximum power points of the
PV system and WECS implemented in the charging station. The
experimental results obtained from the daily operation of the
constructed EV charging station demonstrated that it not only
provides electric energy to charge EVs but also helps the local urban
power distribution network to provide enough electric energy in
response to the load demand, in particular, around 20e23 o’clock
when the load demand peaks in the grid.
References
[1] Fathabadi H. Lambert W function-based technique for tracking the maximum
power point of PV modules connected in various configurations. Renew En-
ergy 2015;74:214e26.
[2] Rezk H, Eltamaly AM. A comprehensive comparison of different MPPT tech-
niques for photovoltaic systems. Sol Energy 2015;112:1e11.
[3] Enslin JHR, Wolf MS, Snyman DB, Swiegers W. Integrated photovoltaic
maximum power point tracking converter. IEEE Trans Ind Electron 1997;44:
769e73.
[4] Park M, Yu In-Keun. A study on the optimal voltage for MPPT obtained by
surface temperature of solar cell. In: 30th annual conference of IEEE, vol. 3;
2004. p. 2040e5.
[5] Masoum MAS, Dehbonei H, Fuchs EF. Theoretical and experimental analyses
of photovoltaic systems with voltage and current-based maximum power-
point tracking. IEEE Trans Energy Convers 2002;17(4):514e22.
[6] Noguchi T, Togashi S, Nakamoto R. Short-current pulse-based maximum-po-
wer-point tracking method for multiple photovoltaic-and-converter module
system. IEEE Trans Ind Electron 2002;49(1):217e23.
[7] Salah CB, Ouali M. Comparison of fuzzy logic and neural network in maximum
power point tracker for PV systems. Electr Power Syst Res 2011;81:43e50.
[8] Algazar MM, AL-monier H, EL-halim HA, El Kotb Salem ME. Maximum power
point tracking using fuzzy logic control. Int J Electr Power Energy Syst
2012;39(1):21e8.
[9] Guenounou O, Dahhou B, Chabour F. Adaptive fuzzy controller based MPPT for
 H. Fathabadi / Energy 132 (2017) 1e11
photovoltaic systems. Energy Convers Manag 2014;78:843e50.
[10] Rizzo SA, Scelba G. ANN based MPPT method for rapidly variable shading
conditions. Appl Energy 2015;145:124e32.
[11] Abdelsalam AK, Massoud AM, Ahmed S, Enjeti PN. High-performance adaptive
perturb and observe MPPT technique for photovoltaic-based microgrids. IEEE
Trans Power Electron 2011;26(4):1010e21.
[12] Jiang JA, Huang TL, Hsiao YT, Chen CH. Maximum power tracking for photo-
voltaic power systems. J Sci Eng 2005;8(2):147e53.
[13] Ahmed J, Salam Z. An improved perturb and observe (P&O) maximum power
point tracking (MPPT) algorithm for higher efficiency. Appl Energy 2015;150:
97e108.
[14] Muthuramalingam M, Manoharan PS. Comparative analysis of distributed
MPPT controllers for partially shaded stand alone photovoltaic systems. En-
ergy Convers Manag 2014;86:286e99.
[15] Tey KS, Mekhilef S. Modified incremental conductance MPPT algorithm to
mitigate inaccurate responses under fast-changing solar irradiation level. Sol
Energy 2014;101:333e42.
[16] Bazzi AM, Krein PT. Concerning maximum power point tracking for photo-
voltaic optimization using ripple-based extremum seeking control. IEEE Trans
Power Electron 2011;26(6):1611e2.
[17] Munteanu I, Bratcu AI. MPPT for grid-connected photovoltaic systems using
ripple-based Extremum Seeking Control: analysis and control design issues.
Sol Energy 2015;111:30e42.
[18] Orabi M, Hilmy F, Shawky A, Qahouq JAA, Hasaneen E-S, Gomaa E. On-chip
integrated power management MPPT controller utilizing cell-level architec-
ture for PV solar system. Sol Energy 2015;117:10e28.
[19] Kotti R, Shireen W. Efficient MPPT control for PV systems adaptive to fast
changing irradiation and partial shading conditions. Sol Energy 2015;114:
397e407.
[20] Bayod-Rújula A-A, Cebollero-Abi
an J-A. A novel MPPT method for PV systems
with irradiance measurement. Sol Energy 2014;109(1):95e104.
[21] Jiang J-A, Su Y-L, Shieh J-C, Kuo K-C, Lin T-S, Lin T-T, et al. On application of a
new hybrid maximum power point tracking (MPPT) based photovoltaic sys-
tem to the closed plant factory. Appl Energy 2014;124:309e24.
[22] Ahmed J, Salam Z. A maximum power point tracking (MPPT) for PV system
using cuckoo search with partial shading capability. Appl Energy 2014;119:
118e30.
[23] Parlak KS. FPGA based new MPPT (maximum power point tracking) method
for PV (photovoltaic) array system operating partially shaded conditions.
Energy 2014;68:399e410.
[24] Chen Y-T, Lai Z-H, Liang R-H. A novel auto-scaling variable step-size MPPT
method for a PV system. Sol Energy 2014;102:247e56.
[25] Daraban S, Petreus D, Morel C. A novel MPPT (maximum power point
tracking) algorithm based on a modified genetic algorithm specialized on
tracking the global maximum power point in photovoltaic systems affected by
partial shading. Energy 2014;74:374e88.
[26] Bouilouta A, Mellit A, Kalogirou SA. New MPPT method for stand-alone
photovoltaic systems operating under partially shaded conditions. Energy
2013;55:1172e85.
[27] Fathabadi H. Comparative study between two novel sensorless and sensor
based dual-axis solar trackers. Sol Energy 2016;138:67e76.
[28] Fathabadi H. Novel high accurate sensorless dual-axis solar tracking system
controlled by maximum power point tracking unit of photovoltaic systems.
Appl Energy 2016;173:448e59.
[29] Fathabadi H. Novel online sensorless dual-axis sun tracker. IEEE/ASME Trans
Mechatron. 2017;22(1):321e8.
[30] Fathabadi H. Harmonic elimination of quasi-sine rotor injected DFIG-based
wind power generation systems connected to electric power networks. Int J
Electr Power Energy Syst 2015;69:399e405.
[31] Fathabadi H. Control of a DFIG-based wind energy conversion system oper-
ating under harmonically distorted unbalanced grid voltage along with non-
sinusoidal rotor injection conditions. Energy Convers Manag 2014;84:60e72.
[32] Dai J, Liu D, Wen L, Long X. Research on power coefficient of wind turbines
based on SCADA data. Renew Energy 2016;86:206e15.
[33] Fathabadi H. Maximum mechanical power extraction from wind turbines
using novel proposed high accuracy single-sensor-based maximum power
point tracking technique. Energy 2016;113:1219e30.
[34] Kumar D, Chatterjee K. A review of conventional and advanced MPPT algo-
rithms for wind energy systems. Renew Sustain Energy Rev 2016;55:957e70.
[35] Castello
J, Espí JM, García-Gil R. Development details and performance
assessment of a wind turbine emulator. Renew Energy 2016;86:848e57.
[36] Lalouni S, Rekioua D, Idjdarene K, Tounzi AM. An improved MPPT algorithm
11
for wind energy conversion system. J Electr Syst 2014;10(4):484e94.
[37] Zhong QH, Ruan Y, Zhao MH, Tan L. Application of variable-step hill climbing
searching in maximum power point tracking for DFIG wind power generation
system. Dianli Xit Baohu yu Kongzhi/Power Syst Prot Control 2013;41(9):
67e73.
[38] Abdullah MA, Yatim AHM, Tan CW, Saidur R. A review of maximum power
point tracking algorithms for wind energy systems. Renew Sustain Energy Rev
2012;16:3220e7.
[39] Dizqah AM, Maheri A, Busawon K, Kamjoo A. A multivariable optimal energy
management strategy for standalone DC microgrids. IEEE Trans Power Syst
2015;30(5):2278e87.
[40] Oh K-Y, Park J-Y, Lee J-S, Lee J. Implementation of a torque and a collective
pitch controller in a wind turbine simulator to characterize the dynamics at
three control regions. Renew Energy 2015;79:150e60.
[41] Urtasun A, Sanchis P, Marroyo L. Small wind turbine sensorless MPPT:
robustness analysis and lossless approach. IEEE Trans Ind Appl 2014;50(6):
4113e21.
[42] Fathabadi H. Novel fast dynamic MPPT (maximum power point tracking)
technique with the capability of very high accurate power tracking. Energy
2016;94:466e75.
[43] Fathabadi H. Novel photovoltaic based battery charger including novel high
efficiency step-up DC/DC converter and novel high accurate fast maximum
power point tracking controller. Energy Convers Manag 2016;110:200e11.
[44] Kazmi SMR, Goto H, Guo HJ, Ichinokura O. A novel algorithm for fast and
efficient speed-sensorless maximum power point tracking in wind energy
conversion systems. IEEE Trans Ind Electron 2011;58(1):29e36.
[45] Fathabadi H. Novel fast and high accuracy maximum power point tracking
method for hybrid photovoltaic/fuel cell energy conversion systems. Renew
Energy 2017;106:232e42.
[46] Fathabadi H. Novel high-efficient unified maximum power point tracking
controller for hybrid fuel cell/wind systems. Appl Energy 2016;183:
1498e510.
[47] Goli P, Shireen W. PV powered smart charging station for PHEVs. Renew
Energy 2014;66:280e7.
[48] Brenna M, Dolara A, Foiadelli F, Leva S, Longo M. Urban scale photovoltaic
charging stations for electric vehicles. IEEE Trans Sustain Energy 2014;5:
1234e41.
[49] Nunes P, Farias T, Brito MC. Day charging electric vehicles with excess solar
electricity for a sustainable energy system. Energy 2015;80:263e74.
[50] Goli P, Shireen W. Wind powered smart charging facility for PHEVs. In: IEEE
energy conversion congress and exposition (ECCE 2014); 2014. p. 1986e91.
[51] Sparacino AR, Grainger BM, Kerestes RJ, Reed GF. Design and simulation of a
DC electric vehicle charging station connected to a MVDC infrastructure. In:
IEEE energy conversion congress and exposition (ECCE 2012); 2012.
p. 1168e75.
[52] Li H, Liu H, Ji A, Li F, Jia Y. Design of a hybrid solar-wind powered charging
station for electric vehicles. In: 2013 international conference on materials for
renewable energy and environment (ICMREE 2013); 2013. p. 977e81.
[53] Fathabadi H. Utilization of electric vehicles and renewable energy sources
used as distributed generators for improving characteristics of electric power
distribution systems. Energy 2015;90:1100e10.
[54] Fathabadi H. Novel high efficiency DC/DC boost converter for using in
photovoltaic systems. Sol Energy 2016;125:22e31.
[55] Rola~n A, Luna A, V
azquez G, Aguilar D. Modeling of a variable speed wind
turbine with a permanent magnet synchronous generator. In: Proc. IEEE in-
ternational symposium on industrial electronics (ISIE’09); 2009. p. 734e9.
[56] Fathabadi H. Novel high efficient speed sensorless controller for maximum
power extraction from wind energy conversion systems. Energy Convers
Manag 2016;123:392e401.
[57] Lingfeng W, Singh C, Kusiak A. Wind power systems: applications of
computational intelligence. Springer; 2010.
[58] Fathabadi H. Novel filter based ANN approach for short-circuit faults detec-
tion, classification and location in power transmission lines. Int J Electr Power
Energy Syst 2016;74:374e83.
[59] Fathabadi H. Two novel proposed discrete wavelet transform and filter based
approaches for short-circuit faults detection in power transmission lines. Appl
Soft Comput 2015;36:375e82.
[60] Fathabadi H. Two novel techniques for increasing energy efficiency of
photovoltaic-battery systems. Energy Convers Manag 2015;105:149e66.
[61] Fathabadi H. Power distribution network reconfiguration for power loss
minimization using novel dynamic fuzzy c-means (dFCM) clustering based
ANN approach. Int J Electr Power Energy Syst 2016;78:96e107.