Scientia Iranica D (2020) 27(6), 3162{3174

Sharif University of Technology

Scientia Iranica
Transactions D: Computer Science & Engineering and Electrical Engineering
http://scientiairanica.sharif.edu

Asymmetrical fuzzy logic control-based MPPT

algorithm for stand-alone photovoltaic systems under
partially shaded conditions
P. Verma , R. Garg, and P. Mahajan
Department of Electrical Engineering, Delhi Technological University, Bawana Road, Delhi, India-110042.
Received 8 September 2018; received in revised form 21 December 2018; accepted 18 February 2019

KEYWORDS Abstract. Partial Shading Conditions (PSCs) in Photovoltaic (PV) system represent
Solar photovoltaic an inevasible situation that curtails the PV array output by exhibiting multiple peaks in
system; its Power-Voltage (P-V) curve. The multiple peaks consist of a single Global Maximum
Maximum power Power Point (GMPP) and many Local Maximum Power Points (LMPPs). The presence
point tracking; of multiple peaks makes tracking of maximum power point quite dicult and demands
Perturb and an ecient controller to track the global peak of the P-V curve. In the present work, a
observation algorithm; novel intelligent Asymmetrical Fuzzy Logic Control (AFLC) based Maximum Power Point
Fuzzy and Tracking (MPPT) algorithm was proposed for tracking GMPP. The fuzzy membership
asymmetrical fuzzy functions of the proposed algorithm were optimized using a heuristic approach. The
algorithm; algorithm was designed, developed, and analyzed using MATLAB/simulink. Furthermore,
Shading loss; to establish the superiority of the proposed AFLC algorithm, it was compared with
DC-DC boost conventional Perturb and Observe (P&O) algorithm and intelligent Fuzzy Logic Control
converter. (FLC) based algorithm for GMPP tracking and shading losses under Standard Test
Condition (STC) and partially shaded conditions.

© 2020 Sharif University of Technology. All rights reserved.

1. Introduction resources, photovoltaic energy conversion is gaining

widespread acceptance due to being green, abundant in
Energy is a necessity for everyday life and due to nature, and relatively cheap [1]. Sun is a massive source
depletion of fossil fuels, this demand may not be of energy which can be converted into a utilizable form
ful lled by conventional power generation systems only. by means of Solar Photovoltaic (SPV) cell.
To overcome this major concern, renewable resources The performance of a SPV system highly depends
such as solar, hydro, wind, hydrogen fuel cell, biogas, on varying meteorological conditions like insolation,
etc. have taken the center stage of the emerging temperature, etc. PhotoVoltaic (PV) panels produce
technology. Amongst these non-conventional energy optimum output under Standard Test Condition (STC)
(irradiation 1 kW/m2 and temperature 25 C) and it
can be negatively a ected by partial shading. Shading
*. Corresponding author. is a condition in which PV array is not uniformly
E-mail addresses: [email protected] (P. Verma)
[email protected] (R. Garg) irradiated. Shading on solar PV array is formed
[email protected] (P. Mahajan) because of dense clouds, nearby buildings, big trees,
towers, dust, and panel aging or cracking. A major
doi: 10.24200/sci.2019.51737.2338 consequence of shading is that it exhibits multiple
 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3163

peaks in Power-Voltage (PV) and Current-Voltage network, ripple correction control current sweep, DC-
(I-V) curves of PV array, which may account for Link capacitor droop control, etc. Ripple correction
conventional Maximum Power Point Tracking (MPPT) control proposed by by Esram et al. [14] enjoyed
algorithms converging to Local Maximum Power Point higher tracking eciency under a uniform shading
(LMPP) instead of Global Maximum Power Point condition and its eciency was signi cantly reduced
(GMPP), resulting in power loss. Hence, mitigating under PSC as it took a longer time to search for
the e ect of partial shading is a major practical GMPP.
challenge. Recently, Balasankar, et al. considered the de-
Several conventional MPPT techniques includ- ployment of P&O under a PSC [15]. In [16], MPPT
ing perturb and observation, incremental conductance algorithm based on an Adaptive Neuro-Fuzzy Inference
(INC), Fractional Open-Circuit Voltage (FOCV), etc. System (ANFIS) was presented. ANFIS model for
were reviewed and addressed in [2,3]. Among these tracking MPP takes temperature and irradiation as
conventional MPPT methods, Perturb and Observe input and duty cycle is the output. In [17], Sahoo
(P&O) is widely used because of its simplicity and detected the presence of PSC by examining the voltage
ease of implementation. However, its reliability is at each PV module; however, this technique required
low under Partial Shading Conditions (PSCs) because an equal number of voltage sensors as the PV module
the perturbation process makes the operating point in the array. Ishaque et al. [18] presented an impressive
lurch around the maximum power point, resulting in method based on Particle Swarm Optimization (PSO)
wastage of power. This challenge can be met by algorithm. However, it is too complicated to be
reducing the step size of perturbation; however, small commercially applied as some parameters need to be set
perturbation size delays the MPPT [4,5]. Al-Majidi by the user. Karami et al. [19] applied Arti cial Neural
et al. proposed a novel MPPT algorithm in [6] in the Network (ANN). The problem faced by ANN-based
performances of P&O and Fuzzy Logic Control (FLC) algorithm is that it is quite dependent on available
based MPPT for the grid-connected PV system were training data under di erent environmental conditions;
compared. The proposed algorithm accurately tracked therefore, the data need to be revised again when array
maximum power with no drift problem. con guration changes. Authors in [20{22] reviewed
INC algorithm o ers improved tracking accu- algorithms such as an arti cial bee, ant colony, PSO,
racy under steady-state and variable environmental ashing re ies, and grey wolf algorithms. Although
conditions. The main advantages provided by INC the genetic algorithm is characterized by better sta-
algorithm over the P&O algorithm were dissected bility, its implementation is a signi cant challenge as
by Liu et al. [7]. The algorithm was formulated it involves complicated calculation and equations and
on the basis that the rise of the PV module power some data need to be set by the user.
curve was zero at the MPP, positive on the left side, In [23], the proposed system tracks maximum
and negative on the right position of the MPP. The power utilizing di erential evolution. The proposed
derivative algorithm was applied to nd the MPP. This technique appear to have a better response regarding
method puts a considerable computation burden on the fast convergence in both uniform and non-uniform
controller because the di erentiation process involves shading conditions. It also eliminates uctuations
a relatively complex decision-making procedure [8,9]. around MPP. Inputs such as insolation level and
Therefore, implementing INC algorithm increases the the temperature to SPV are ephemeral. Therefore,
cost and complexity of the system and it may produce control algorithms need to be implemented to cope
unsatisfactory results at a low-insolation level. Further, with the uncertainty in the quantities of interest. FLC
Reisi et al. [10] distinguished MPPT algorithms into technique, a soft computing tool, can deal with these
two categories: direct and indirect methods. Indirect uncertainties [24]. FLC can incorporate a conventional
MPPT algorithms such as FOCV and short-circuit design to track MPP under ambient conditions, as
current cannot trace MPP in any given environmen- suggested by Sundareswarm, et al. [25]. Verma, et
tal condition, given that these methods require prior al. [26] compared P&O with intelligent FLC algorithm
knowledge of P-V characteristics. Direct method [11] under a partially shaded condition for stand-alone
based algorithms do not require prior knowledge of PV systems and simulation results illustrated that
P-V characteristics and can track MPP under any FLC could track maximum power with better tracking
weather conditions. Authors in [12] compared the eciency. Results also illustrated that conventional
e ects of di erent membership functions of FLC. MPPT might track LMPP instead of GMPP under
Triangular membership function produced the best dynamic conditions.
result, while Gaussian membership function generated The present work focuses on the tracking of
poor performance in the considered cases. Asim et global peak out of multiple peaks under PSC. Solar
al. [13] carried out performance analysis for MPPT photovoltaic panels under PSCs are not uniformly
by di erent controllers such as P&O, FLC, neural irradiated, which makes P-V characteristic exhibit
3164 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

more than one peak. Out of these multiple peaks, 2. Solar PV cell
there is only one peak exhibiting maximum power,
i.e., GMPP. The conventional algorithm may not Solar cell works on the principle of PV e ect. It is an
be able to track GMPP and lurch around LMPP. active transducer that converts energy from sunlight
To overcome the drawbacks of a conventional al- (photons) into electricity (current). The voltage pro-
gorithm, FLC was implemented so as to enhance duced by the solar cells is minimal (0.5 V to 0.8 V)
the performance of MPPT by tracking GMPP. The to achieve the desired output so that they can remain
MPPT tracking eciency was further enhanced using connected in series. Solar cells connected in series
a novel, intelligent asymmetrical fuzzy logic controller- make PV module and PV module connected in a series-
based algorithm. For asymmetrical FLC, the au- parallel con guration, as per power requirement, makes
thors optimized the universe of discourse by ne- PV array [27]. Figure 2 presents the ideal and practical
tuning it at the center for e ective implementation equivalent circuits of the single-diode model used in the
of MPPT. The proposed asymmetrical FLC MPPT proposed system.
algorithm distinguishes between GMPP and LMPP The output current Io=p of PV cell is given in
and contributes towards global peak tracking. By Eq. (1):
using an asymmetrical fuzzy controller, the tracked
power is higher than other conventional algorithms, Io=p =Ics Isat
thus proving the superiority of the proposed algorithm.    
Io=p  Rs )
Moreover, shading losses are reduced using Asymmet-  exp q(Vo=pN+AkT 1
rical Fuzzy Logic Control (AFLC) algorithm. The s ac
proposed MPPT algorithm excelled in both steady
state and dynamic conditions. The relatively new idea Vo=p + Io=p Rs
; (1)
of asymmetrical FLC-based MPPT algorithm under Rsh
PSCs for stand-alone PV systems can be applied to where Io=p is the output current of PV cell, Ics
large solar farms. These studies shall be helpful for is the photon current, Isat is the diode saturation
system designers.
1.1. System con guration
Figure 1 presents the block diagram of a stand-alone
PV system. To track maximum available power in
a given environmental condition, MPPT technique is
utilized by adjusting the duty ratio of a boost con-
verter. The proposed asymmetrical FLC-based MPPT
algorithm was designed, developed, and validated to
track the global peak from multiple peaks of the P-V
curve under partially shaded conditions, which may not
be eciently traced by other conventional approaches.
The components of the proposed system are explained Figure 2. Equivalent circuit of a single-diode model of a
in the following sections. Photovoltanic (PV) cell.

Figure 1. Block diagram of a stand-alone Photovoltanic (PV) system.

 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3165

current, q is the elementary charge (1.602 10 19 C), and change in irradiation causes appreciable variations
Vo=p is the output voltage of a PV cell, Ns is the in current. Also, according to Figure 3, when modules
number of series cell, A is the ideal factor of the are uniformly irradiated, there is only one maximum.
cell dependent on PV technology, K is the Boltzmann
constant (1:38 10 23 J), and Tac is the actual operating 3. DC-DC converter
temperature.
In the present work, sun power solar panel X 22 In the proposed PV system, the boost converter is used
360 PV module is used. Table 1 shows the sun as a dc-dc converter (Figure 1). The main components
power solar panel speci cations used for simulating of the boost converter along with IGBT switch are the
and modeling PV module. An array of (3  1) PV series inductor and shunt capacitor, which are passive
modules viz. M1, M2, M3, each characterized by components. The values of an inductor, capacitor, duty
360 W, connected in series is considered, as shown in ratio, and resistive load are calculated below, as given
Figure 1, and is simulated using MATLAB/simulink. by Eqs. (2){(5).
The maximum power obtained by the array is (360  3) V
1.08 kW. L = o=p ; (2)
(
I1 fsw )
Figure 3 shows the e ect of irradiation and  
temperature on the I-V and P-V characteristics of Va
PV array. I-V and P-V curves show that with a =1 ; (3)
Vo=p
change in temperature, voltage changes appreciably
Ia
C= ; (4)
Table 1. Speci cation of sun power Photovoltanic (PV) (
V fsw )
Module. Rin
SPR-X22-360 Ro = ; (5)
(1 )2
Power nominal (Max), Pnom 360 W where
l1 is the output ripple current and is considered
Rated voltage, Vm 60.6 V
as 10% of the input current, Vo=p is the output voltage,
fsw is the switching frequency, is the duty ratio, Va is
Rated current, Im 5.94 A the input voltage, and Ia is the average output current.

V is the peak ripple voltage whose value is taken as
Open-circuit voltage, Voc 69.5 V 3% of the output voltage, Rin is the input resistance,
Short-circuit current, Iscc 6.48 A and Ro is the load resistance which appears to be 122 .
The values of inductance, capacitance, and duty ratio
Total no. of cell in series, Ns 96 are 1.1 mH, 500 F, and 0.5{0.7, respectively.

Figure 3. Current vs. voltage and power vs. voltage curves of the proposed Solar PV (SPV) system with (a) variable
irradiation and (b) variable temperature.
3166 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

4. MPPT techniques 4.2. Intelligent algorithm (AFLC)

4.1. Conventional algorithm (P&O algorithm) Fuzzy logic is a logical system that does not require an
The P&O maximum power point algorithm works on accurate mathematical model. FLC uses \If...then..."
the condition that on the right side of the MPP, if command to frame a rule base. The fuzzy inference
voltage increases, power decreases and perturbation system can be Mamdani or Sugeno, and both can have
is made on the opposite direction, whereas towards symmetrical or asymmetrical membership functions
the left side of MPP, if voltage increases, power also [28,29]. The main drawback of conventional fuzzy
increases; then, perturbation is made in the same logic algorithms is that they may not track GMPP.
direction. Implementation of a P&O algorithm in the In the proposed work, the asymmetrical FLC-based
proposed PV system is shown Figure 4. control algorithm was designed. The input/output
The main drawback of the P&O algorithm is its membership functions of symmetrical FLC were further
inability to track maximum power point due to the os- ne-tuned by the heuristic approach with a carefully
cillations near MPP region under varying environmen- designed rule base to track MPP under PSCs.
tal conditions. To overcome these disadvantages, FLC- In the proposed AFLC algorithm, the asymmetric
based algorithm is implemented that can minimize distribution of input and output membership functions
oscillations near the operating point; hence, energy includes both convergent (+ ) and divergent ( )
wastage in a PV system is also minimized. Further, types of asymmetry for the fuzzy variables. The
in the proposed AFLC-based algorithm, the output values of [+ , ] were tuned to achieve MPP without
power is further enhanced by asymmetrical distribution oscillations. Figure 5 shows the block diagram of the
of fuzzy membership functions. asymmetrical FLC-based control algorithm for MPPT.

Figure 4. Block diagram of the P&O algorithm.

Figure 5. Block diagram of asymmetrical Fuzzy Logic Control (FLC)-based control algorithm for Maximum Power Point
Tracking (MPPT).
 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3167

First, fuzzi cation, inference method, and defuzzi ca- B1 (Big 1), B2 (Big 2), and B3 (Big 3) [24]. In
tion are elaborated below. the proposed AFLC, the membership functions are
denser at the centre to provide greater sensitivity in the
4.2.1. Fuzzi cation region near the MPP. Input membership functions are
In the proposed asymmetrical FLC-based MPPT algo- normalized and suitable tuning gains are used to match
rithm, two inputs include error and change in error. the inputs to the respective universe of discourse.
The power is calculated based on the sensed voltage
and current. Error is then calculated using Eq. (6).
Change in error is derived from the derivative of error 4.2.2. Inference method
and can be calculated through Eq. (7): The inference method applies the rules to the fuzzy
input to determine the fuzzy output. Rules are made
Po=p(n) Po=p(n 1) based on the membership function of error and change
Er (n) = ; (6) in error. In the proposed system, the rules have been
Io=p(n) Io=p(n 1)
made using \If...then..." logic. For example, if an error

Er (n) = Er (n) Er (n 1) ; (7) is S1 and change in error is S3, then the duty ratio is
ZO.
where Er is the calculated error, Po=p and Io=p are the The idea behind making a rule base in the form
power and current output of the proposed PV system, of a matrix is to bring the operating point closer to
and
Er is the change in error. the maximum power point with less uctuations by
The universe of discourse or membership function increasing or decreasing the duty ratio as per the
is a curve that denotes how each point in the input direction in which maximum peak occurs [30,31]. In
space is mapped in between 0 and 1. Input and output Figure 7, the rule base for the proposed system is shown
universes of discourse of fuzzy sets consist of seven which consists of 49 fuzzy control rules. These rules can
triangular membership functions. Figure 6 shows the also be represented in a 3D graph known as a surface
asymmetrical distribution of fuzzy input and output viewer, as shown in Figure 8. In the present work,
membership functions used in the proposed system Mamdani's Max-Min method is used for inference of
for tracking maximum power in a steady state and fuzzy controller. The output membership function of
a dynamic condition. The universe of discourse for each rule is given by the minimum operator, whereas
the input variable is divided into seven fuzzy sets: S1 collective fuzzy output (x) is provided by maximum
(Small 1), S2 (Small 2), S3 (Small 3), ZO (Zero), operator [32].

Figure 6. Membership functions for input variable `error', error change', and output variable `duty ratio' for
asymmetrical Fuzzy Logic Control (FLC).
3168 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

asymmetrical FLC-based MPPT algorithm is employed

to track maximum available power under varying
environmental conditions. The result of AFLC has
been compared with those of P&O and FLC under
three conditions: (i) STC, (ii) PSC, and (iii) dynamic
conditions.
Moreover, the performance of the proposed algo-
rithm has been evaluated and compared with respect
to other algorithms on the basis of performance criteria
viz. GMPP tracking and shading losses:
Figure 7. Fuzzy rule base for computing the output (i) Maximum power: The maximum power point is
variable called `duty ratio' in the matrix form.
the point with maximum voltage and current. It
can be determined by Eq. (10):
PMP = VMP  IMP : (10)
(ii) Shading losses: In spite of technological ad-
vancement, the adverse e ect of partial shading
on the PV system results in power loss. The
power loss due to shading is called shading loss.
Shading loss is the di erence in power between the
maximum power obtained from an array under
STC (PMP ;without shading ) and the total maximum
available power under PSCs (PMP; shading ) [33].
It can be represented by Eq. (11):
PMP ;shading losses =PMP ;without shading
Figure 8. 3D surface view representation between two
inputs (error and change in error) and output (duty ratio) PMP ; shading : (11)
generated by Asymmetrical Fuzzy Logic Control (AFLC) In the present study, all the three PV modules of
of the proposed Photovoltanic (PV) system. the proposed system are connected in series and
the maximum power obtained from this series ar-
4.2.3. Defuzzi cation rangement is 1.08 kW. The modules are excited at
Defuzzi cation is used to convert the fuzzy inference di erent irradiation levels to give various shading
output into crisp output which can be obtained through patterns, as shown in Figure 9.
Eq. (8): The following cases are considered:
zo = defuzzifier (x); (8)
where x is the aggregate output and defuzzi er is the
Case 1: All the three PV modules (M1, M2, M3)
are at STC, i.e., solar insolation at 1000 W/m2 and
defuzzi cation operator. temperature at 25 C. Case 1 is shown in Figure 9(a);
Defuzzi cation can be done by using the center
of area, the max criterion method, etc. In the present
work, the center of the area defuzzi er operator is used Case 2: PV module M1 is at 1000 W/m2 , M2 at
and represented by Eq. (9): 900 W/m2 , and M3 at 500 W/m2 . Case 2 is shown in
Pn Figure 9(b);
x= P i=1  (xi ) xi ; (9)
n
i=1 xi Case 3: PV module M1 is at 500 W/m2 , M2 at 400
where xi is the activation degree on rule `i', xi is W/m2 , and M3 at 700 W/m2 . Case 3 is shown in
the center of the Max-Min composition of the output Figure 9(c);
membership functions, and x is the required output,
i.e., duty ratio. Case 4: PV module M1 is at 450 W/m2 , M2 at 250
W/m2 , and M3 at 500 W/m2 . Case 4 is shown in
5. Simulation results and performance Figure 9(d);
evaluation
Case 5: PV module M1 is at 100 W/m2 , M2 at 100
The MATLAB/simulink environment is used to de- W/m2 , and M3 at 200 W/m2 . Case 5 is shown in
velop a stand-alone PV system and the proposed Figure 9(e).
 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3169

Table 2. Power at Global Maximum Power Point (GMPP) and Local Maximum Power Point (LMPP) under various
shading patterns.
Case 1 Case 2 Case 3 Case 4 Case 5
PGMP P (W) 1080 630 425 250 78
PLMP P (W) { 332; 555 227; 347 141; 198 55

Figure 9. Shading patterns of Photovoltaic (PV) modules for the proposed system.

Figure 10. Simulated current vs. voltage and power vs. voltage curves for the di erent shading patterns of the proposed
system.
Figure 10 shows the I-V and P-V characteristics of FLC algorithms in di erent shading patterns. The
the proposed stand-alone PV system in di erent shad- results are tabulated in Table 3. Table 4 gives the
ing patterns considered. Table 2 gives the global and lo- corresponding shading losses.
cal maximum power rates in di erent shading patterns.
Discussion of the considered cases are as follows:
5.1. Steady-state response
Figure 11 compares the steady-state response of the Case 1: In this case, the highest maximum power is
proposed AFLC-based MPPT with those of P&O and tracked by AFLC algorithm, which is 970.5 W, with
3170 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

Figure 11. Simulated power vs. time plot of the three algorithms under the steady-state condition with di erent shading
patterns.
 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3171

Table 3. Steady-state response of various Maximum Power Point Tracking (MPPT) techniques under study.
MPPT Case 1 Case 2 Case 3 Case 4 Case 5
techniques
PMP (W) 930 540 402 201 46
P&O V (V) 360 276 236.5 167 82.8
I (A) 2.583 1.956 1.699 1.203 0.55

PMP (W) 965.4 578.5 417.5 220.5 49.5

FLC V (V) 360 279.4 205 161.5 81.06
I (A) 2.681 2.070 2.036 1.362 0.61

PMP (W) 970.5 583.5 419 223 50.5

AFLC V (V) 360 280.8 204 172.5 80.25
I (A) 2.695 2.080 2.053 1.380 0.62

Table 4. Shading loss (W) in various Maximum Power Point Tracking (MPPT) techniques under study.
MPPT Case 1 Case 2 Case 3 Case 4 Case 5
techniques
P&O 150 540 678 879 1034
FLC 114.6 501.5 662.5 859.5 1030.5
AFLC 109.5 496.5 661 857 1029.5

no uctuations around GMPP. The shading losses are

the lowest viz. 109.5 W. FLC tracks 965.4 W, while
conventional P&O MPPT algorithm tracks only 930 W
power, i.e., the lowest power and shading losses are the
highest at 150 W;
Case 2: In this case, the proposed AFLC-based MPPT
algorithm tracks the maximum power of 583.5 W and
the shading losses of 496.5 W are least among the values
obtained by other MPPT approaches. Moreover, in this
case, the AFLC MPPT algorithm tracks power more Figure 12. Comparative analysis of tracked power on the
accurately than other approaches; bar chart.

Case 3: In this case, the highest maximum power tracked by AFLC algorithm which is 50.5 W, with no
is tracked by AFLC algorithm which is 419 W, with uctuations around GMPP. The shading losses are also
no uctuations around GMPP. The shading losses the lowest viz. 1029.5 W. FLC tracks 49.5 W and con-
are the lowest viz. 661 W. FLC tracks 417.5 W and ventional P&O MPPT algorithm tracks 46 W power,
Conventional P&O MPPT algorithm tracks 402 W i.e., the lowest power tracked with large perturbations.
power, i.e., the lowest power tracked and shading losses Shading losses are the highest at 1034 W.
are also the highest at 678 W;
According to Tables 3 and 4, the power tracked
Case 4: In this case, the highest maximum power is maximum and shading losses are minimum in the
is tracked by AFLC algorithm which is 223 W, with proposed asymmetrical FLC. In addition, according to
no uctuations around GMPP. The shading losses are Figure 11, it can be seen that P&O algorithm has large
also lowest viz. 857 W. FLC tracks 220.5 W and perturbations in the output. Figures 12 and 13 show
the conventional P&O MPPT algorithm tracks 201 W the comparative analysis of tracked powers and shading
power, i.e., the lowest power tracked in this case and losses on the bar chart, respectively.
shading loss is high which is 879 W;
5.2. Transient response
Case 5: In this case, the highest maximum power is The proposed asymmetrical FLC-based MPPT has
3172 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

is maximum and shading losses are minimum in the

proposed asymmetrical FLC-based algorithm. In ad-
dition, the settling time of the proposed asymmetrical
FLC is the shortest in all conditions, while P&O takes
the longest settling time.

6. Conclusion
This paper proposed an intelligent asymmetrical Fuzzy
Logic Control (FLC)-based Maximum Power Point
Figure 13. Comparative analysis of shading loss on the Tracking (MPPT) algorithm for the stand-alone Pho-
bar chart. tovoltanic (PV) system in partial shading conditions.
To establish the superiority of the proposed algo-
rithm, it was compared with other conventional and
Table 5. Shading patterns of Photovoltaice (PV) intelligent techniques viz. Petrob and Observe (P&O)
Modules for transient response. and FLC. The proposed asymmetrical FLC algorithm
Shading Irradiation Time was designed, developed, and validated to track the
pattern level (kW/m2 ) (sec) global maximum power under various shading scenarios
SP1 [M 1 : 0:9; M 2 : 0:7; M 3 : 0:6] 0 2:5 including steady and dynamic states. Simulation
SP2 [M 1 : 0:5; M 2 : 0:8; M 3 : 0:3] 2:5 5 results demonstrated that the asymmetrical FLC could
SP3 [M 1 : 0:6; M 2 : 0:3; M 3 : 0:1] 5 7 e ectively track the global maximum power point
under various test conditions. Moreover, compared
to other algorithms, the proposed asymmetrical FLC-
been evaluated in transient conditions. The change in based MPPT algorithm had less shading losses and
the insolation level for each PV module in a given time took the shortest settling time to perform. The
duration is shown in Table 5. Figure 14 shows the implementation of the asymmetrical FLC-based MPPT
comparative analysis of the asymmetrical FLC-based algorithm improved the overall steady state and dy-
MPPT algorithm and other approaches in transient namic behavior of the PV system under consideration.
conditions. Tables 6 and 7 show the results and the Therefore, there was low wastage of environment-
corresponding shading losses, respectively. friendly solar power. Furthermore, these studies should
According to Tables 6 and 7, the power tracked be useful to system designers.

Figure 14. Simulated power vs. time plot of the three algorithms under transient conditions.
 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174 3173

Table 6. Transient response of various Maximum Power Point Tracking (MPPT) techniques under study.
0{2.5 (sec) 2.5{5 (sec) 5{7 (sec)
MPPT Power Settling Power Settling Power Settling
techniques (W) time (sec) (W) time (sec) (W) time (sec)
P&O 486 0.26 249 0.6 190 0.2
FLC 517.5 0.19 250.1 0.3 196.5 0.1
AFLC 521.5 0.19 250.6 0.3 198.1 0.1

Table 7. Shading loss (W) in various Maximum Power Point Tracking (MPPT) techniques under study.
MPPT 0{2.5 (sec) 2.5{5 (sec) 5{7 (sec)
techniques Shading loss (W) Shading loss (W) Shading loss (W)
P&O 594 831 890
FLC 562.5 829.9 883.5
AFLC 558.5 829.4 881.9

References novel multi-model neuro-fuzzy-based MPPT for three-

phase grid-connected photovoltaic system photovoltaic
1. Ellabban, O., Abu-Rub, H., and Blaabjerg, F. \Re- system", Sol. Energy, 84(12), pp. 2219{2229 (2010).
newable energy resources: current status, future
prospects and their enabling technology", Renew. Sus- 10. Reisi, A.R., Moradi, M.H., and Jamasb, S. \Classi ca-
tain. Energy Rev., 39(C), pp. 748{764 (2014). tion and comparison of maximum power point tracking
techniques for photovoltaic system", Renew. Sustain.
2. Abdourraziq, M.A., Maarou , M., and Ouassaid, M. Energy Rev., 19, pp. 433{443 (2013).
\A new variable step size INC MPPT method for PV 11. Tsai, H.F. and Tsai, H.L. \Implementation and ver-
systems", Int. Conf. Multimed. Comput. Syst. Proc., i cation of integrated thermal and electrical models
55(7), pp. 1563{1568 (2014). for commercial PV modules", Sol. Energy, 86(1), pp.
3. Koutroulis, E. and Blaabjerg, F. \A new technique for 654{665 (2012).
tracking the global maximum power point of PV arrays 12. Ali, O.A.M., Ali, A.Y., and Sumait, B.S. \Comparison
operating under partial-shading conditions", IEEE J. between the e ects of di erent types of membership
Photovoltaics, 2(2), pp. 184{190 (2012). functions on fuzzy logic controller performance", Int.
4. Patel, H. and Agarwal, V. \Maximum power point Journal of Emerging Engg. Research and Tech., 3(3),
tracking scheme for PV systems operating under par- pp. 76{83 (2015).
tially shaded conditions", IEEE Trans. Ind. Electron, 13. Asim, N., Sopian, K., Ahmadi, S., et al. \A review on
55(4), pp. 1689{1698 (2008). the role of materials science in solar cells", Renewable
Sustainable Energy Rev., 16(8), pp. 5834{5847 (2012).
5. Ji, Y.H., Jung, D.Y., Kim, J.G., et al. \A real max-
imum power point tracking method for mismatching 14. Esram, T., Kimball, J.W., Krein, P.T., et al. \Dynamic
compensation in PV array under partially shaded maximum power point tracking of photovoltaic arrays
conditions", IEEE Trans. Power Electron., 26(4), pp. using ripple correlation control", IEEE Trans. Power
1001{1009 (2011). Electron., 21(5), pp. 1281{1291 (2006).
6. Al-Majidi, S.D., Abbod, M.F., and Al-Raweshidy, H.S. 15. Balasankar, R., Arasu, G.T., and Christy Mano Raj,
\A novel maximum power point tracking technique J.S. \A global MPPT technique invoking partitioned
based on fuzzy logic for photovoltaic systems", Int. J. estimation and strategic deployment of P&O to tackle
of hydrogen Energy, 43(31), pp. 14158{14171 (2018). partial shading conditions" , Sol. Energy, 143, pp. 73{
85 (2017).
7. Liu, C., Wu, B., and Cheung, R. \Advanced algorithm 16. Kharb, R.K., Shimi, S.L., Chatterji, S., et al. \Mod-
for MPPT control of photovoltaic systems", In: Pro- eling of solar PV module and maximum power point
ceedings of the Canadian Solar Buildings Conference, tracking using ANFIS", Renew. Sustain. Energy Rev.,
Montreal; August, pp. 20{24 (2004). 33, pp. 602{612 (2014).
8. Azab, M.A. \New maximum power point tracking for 17. Sahoo, S.K. \Solar photovoltaic energy progress in
photovoltaic systems", International Journal Electrical India: a review", Renew. Sustain. Energy Rev., 59,
Electron. Engineering, 2(8), pp. 1600{1603 (2009). pp. 927{939 (2016).
9. Chaouachi, A., Kamel, R.M., and Nagasaka, K.A. 18. Ishaque, K., Salam, Z., and Syafaruddin \A compre-
\Novel multi-model neuro-fuzzy-based MPPT for hensive MATLAB simulink PV system simulator with
3174 P. Verma et al./Scientia Iranica, Transactions D: Computer Science & ... 27 (2020) 3162{3174

partial, shading capability based on two-diode model", 30. Kuo, T.J. and Chen, J.F. \Novel maximum-power-
Sol. Energy, 85(9), pp. 2217{2227 (2011). point tracking controller for photovoltaic energy con-
19. Karami, N., Moubayed, N., and Outbi, R. \General re- version system", IEEE Trans. Ind. Electron., 48(3),
view and classi cation of di erent MPPT techniques", pp. 594{601 (2001).
Renew. Sustain. Energy Rev., 68(1), pp. 1{18 (2017). 31. Casadei, D. \Single-phase single-stage photovoltaic
20. Li, G., Jin, Y., Akram, M.W., et al. \Application generation system based on a ripple correlation control
of bio-insopired algorithms in maximum power point maximum power point tracking", IEEE Trans. Energy
tracking for PV system under partial shading condi- Convers., 21(2), pp. 562|568 (2006).
tions", A review. Renewable and Sustainable Energy,
81(1), pp. 840{873 (2018). 32. Gupta, N. and Garg, R. \Tuning of asymmetrical
21. Singh, N. \A modi ed variant of grey wolf optimizer", fuzzy logic control algorithm for SPV system connected
Int. Journal of Science & Tech., 27(3), pp. 1450{1466 to grid", International Journal of Hydrogen Energy,
(2018). DOI: 10.24200/SCI.2018.50122.1523 42(26), pp. 16375{16385 (2017).
22. Vaez, S.R.H. and Minaei, Z. \Pulse extraction 33. El-Dein, M.S., Kazerani, M., and Salama, M.M.A.
of pulse like ground motions based on particle \Optimal photovoltaic array recon guration to reduce
swarm optimization algorithm", Int. Journal of Sci- partial shading losses", IEEE Trans. Sustain. Energy,
ence & Tech., 27(1), pp. 134{158 (2018). DOI: 4(1), pp. 145{153 (2013).
10.24200/SCI.2018.5607.1374
23. Ram, J.P., Babu, T.S., and Rajasekar, N. \A com-
prehensive review on solar PV maximum power point Biographies
tracking techniques", Renew. Sustain. Energy Rev.,
67, pp. 826{847 (2017). Pallavi Verma received BTech degree in 2009 from
24. Kumar, P. and Mahajan, A. \Soft computing tech- K.I.E.T., Ghaziabad, India in Electrical and Electron-
niques for the control of an active power lter", IEEE ics Engineering and ME degree in Instrumentation &
Trans. Power Deliv., 24(1), pp. 452{461 (2009). Control in 2015 from NITTTR, Chandigarh, India.
25. Sundareswarm, K., Sankar, P., Nayak, P.S.R., et Presently, she is pursuing her PhD at Electrical Engi-
al. \Enhanced energy output from a PV system un- neering Department at Delhi Technological University,
der partial shaded conditions through arti cial bee Delhi. Her research interests are in renewable energy,
colony", IEEE Trans. Sustain. Energy, 6(1), pp. 198{ power electronics, and control system.
209 (2015).
26. Verma, P., Mahajan, P., and Garg, R. \Comparison Rachana Garg (IEEE SM'10) received the BE and
of intelligent and conventional MPPT algorithms for ME degrees in 1986 and 1989, respectively, from
photoVoltaic system under partially shaded condi- National Institute of Technology, Bhopal, India. She
tions", IEEE International Conference, RDCAPE- obtained her PhD in Electrical Engineering from Delhi
2017, India, pp. 505{510 (2017). University, India in 2009. Presently, she is working as a
27. Gow, J.A. \Development of a model for photovoltaic Professor at Delhi Technological University, Delhi, In-
arrays suitable for use in simulation studies of solar dia. Her areas of interests are modeling of transmission
energy conversion systems", 6th Int. Conf. Power
Electron. Variable Speed Drives, Nottingham, UK, pp. lines, power system operation and control, smart grid,
69{74 (1996). and renewable energy.
28. Al-Gizi, A., Al-Chlaihawi, S., Louzazni, M., et al. \Ge-
netically optimization of an asymmetrical fuzzy logic Priya Mahajan received BE from Thapar Institute
based photovoltaic maximum power point tracking of Engineering and Technology in 1996 and ME from
controller", Advances in Elec. and Computer Engg., Punjab Engineering College in 1998, Punjab, India.
17(4), pp. 69{76 (2017). She has obtained her PhD in Electrical Engineering
29. Liu, C.L., Chen, J.H, Liu, Y.H., et al. \An asymmet- from Delhi University, India in 2015. Presently, she is
rical fuzzy-logic-control-based MPPT algorithm for working as a Professor at Delhi Technological Univer-
photovoltaic systems", Energies, 7, pp. 2177{2193 sity, Delhi, India. Her areas of interests are railway
(2014). electric traction system and power system.