ISSN 0003-701X, Applied Solar Energy, 2018, Vol. 54, No. 4, pp. 235–245. © Allerton Press, Inc., 2018.

DIRECT CONVERSION OF SOLAR

ENERGY INTO ELECTRICAL ENERGY

Comparative Study of Different DC/DC Power Converter

for Optimal PV System Using MPPT (P&O) Method1
Abdellahi Baa, b, *, Chighali Ould Ehsseinb, Mouhamed El Mamy Ould Mouhamed Mahmoudb,
Ouafae Hamdouna, and Aroudam Elhassena
aAbdelmalek
Essaadi University, Tétouan, 93002 Morocco
b
Unité de Recherche en Systèmes Electromécanique, DGEM, ISET-Rosso, Mauritanie
*e-mail: [email protected]
Received September 28, 2017

Abstract—The objective of this work is to show when and how to choose the type of DC/DC power converter
that must be introduced in a stand alone PV solar system’s optimization process, which is necessary to char-
acterize the area’s climatic conditions of the system implementation, take into account the partial shading
effect on the Photovoltaic Generator (PVG). In addition, determine the resistivity of the load that envisaged
to be introduced with the help of the direct coupling responses of the PVG. Based on the MATLAB/Simulink
software and the PVG data sheet choice, we could plot the current-voltage (I-V) and power-voltage (P-V)
characteristic, using the manufacturer data and determining the maximum power point (MPP), as a function
of the area’s weather conditions variations, at the same time observing the direct coupling response of the
load adapted to the PVG. From this observation, we can specify the DC/DC power converter type to be intro-
duced in the production optimization process of a standalone PV solar system. Thus, we must use a Boost
converter type, if the direct coupling voltages responses are higher than those of MPP responses, or a Buck
converter type in the opposite case, or both at the same time Buck-Boost converter type, if we need to
increase and decrease the voltage.

Keywords: PVG, DC/DC power converter, MPPT, Direct coupling, resistive load, shading effect
DOI: 10.3103/S0003701X18040047

INTRODUCTION these techniques has these advantages and disadvan-

The use of the power electronics systems has been tages as it is detailed in the investigation work by [2], in
always solicited in the renewable energies production, this work the author shows the classification and per-
especially those of solar PV systems. In the last six formance of each MPPT technique applied in partial
decades there are more than 500 models of DC/DC shading effect conditions cases. In addition, he was
power converters used in different fields of power elec- able to classify these MPPT techniques into four cate-
tronics application including the renewable energies gories, but what interests us is the last category; the
production [1]. classification according to the DC/DC power con-
PV solar systems are non-linear systems and too verters type to be introduced to act as an impedance
sensitive to the variations of weather conditions, the adapter.
characteristics of the magnitudes delivered by this sys- The DC/DC power converter is the important ele-
tem (current, voltage) have a single point where the ment in PV solar system’s optimization, as it acts as an
power can be maximum. Sometimes it is difficult that interface between the PVG and the adapted load, this
the operating point between the load and the PVG is DC/DC power converter is controlled through its
the optimal point of the system, in this case it becomes switch by a signal generated by the MPPT technique
necessary to introduce the maximum power point called the duty cycle. Practically one of these three
tracking technique (MPPT). This technique is used as types of converter is used in the literature; Buck, Boost
the controller of a DC/DC power converter through and Buck-Boost power converter which are consid-
its switch that acts as an impedance adapter. ered as converters not isolated from the source [3], or
In the literature there are several techniques of on the other hand other DC/DC power converters
maximum power point research, and that each of isolated from the source are used [4].
In this article, we show how to choose the type of
1 The article is published in the original.
converters that must be introduced in the production

235
236 BA et al.

Irradiation IPV Icov

DC/DC
power
PVG VPV converter Vcov R load
K
Temperature

Duty cycle
MPPT

Fig. 1. Description of the system optimized.

optimization of standalonePV solar systems according ⎡ q ⎛⎜VPV + RsIPV ⎞⎟ −1 ⎤

to the adapted load and the weather conditions varia- I PV = N pI ph − N pI rs ⎢⎣e ⎝ AKTN s ⎠ ⎥⎦
tions takin into account the partial shading effect con- (2)
⎛V + RsI PV ⎞
ditions to achieve a good impedance matching. − N pq ⎜ PV ⎟,
⎝ RsRsh ⎠
where:
DESCRIPTION OF THE SYSTEM IPV and VPV are respectively the current and the
The stand alone PV solar system is composed by voltage of the PV cell;
three essential parts; the PVG that is sensitive to the Iph is the photocurrent of the PV cell; it is propor-
weather conditions, a DC/DC power converter and tional to the solar irradiation parameter;
the MPPT technique which play the role of imped- Id and Ish are respectively current of the diode and
ance matching (optimization), and finally we have a that of theparallel resistor;
resistive load (R), and this load can be not a resistive Irs is the diode saturation current;
but inductive or capacitive load (Fig. 1).
Ns and Np are respectively the number of the cells
connected in series and in parallel;
The PVG Modeling Rs and Rsh are respectively the series and parallel
resistances of the PV cell;
The PVG is a module made up by PV cells con-
(q = 1.602 × 10–19 C) the electron charge;
nected together, in series or in parallel or both at the
same time, so it is enough to model a PV cell to deduce (K = 1.38065 × 10–23 J/K) Boltzmann constant;
the complete modeling of the PVG. Figure 2 shows the A is the ideality factor of the diode (in general, it is
electrical modeling of a PV cell [5]. equal to 1 or 2), T is the absolute temperature of the
PV cell in °C.
From this circuit, applying Kirchhoff’s law, we
From the data in Table 1, with the help of the soft-
shall derive the following equations:
ware used, we were able to plot the (V-I) and (V-P)
PVG characteristics, Fig. 3.
I PV = I ph − I d − I sh, (1)

Optimization of the Stand Alone PV Solar System

Rs
IPV The production optimization of the PV solar sys-
tem is based essentially on the maximum power point
Iph Id Ish
tracking technique (MPPT). in the literature we find
D Rsh VPV quite a lot of different techniques used to solve the
solar PV systems optimization problem; Perturb and
Observe (P&O) [6], Incremental Conductance (IC)
[7, 8], Fuzzy logic controller (FLC) [9] and Neural
Networks (NN) technique [10] are often used in the
Fig. 2. Electrical model of a PV cell. optimization process.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

 COMPARATIVE STUDY OF DIFFERENT DC/DC 237

In this article, we would like to show that the opti- Table 1. Datasheet of the Luxor LX-190M module
mization of the PV solar system production is not only Peak Power [PMPP] 190 [WP]
limited to the good performance of the MPPT tech-
nique, but also the choice of a DC/DC power con- Rated Voltage [VMPP] 36.14 [V]
verter type is more important. In addition, this choice
must not be hazardous. Rated Current [IMPP] 5.26 [A]
In solar PV system optimization, the DC/DC
power converter is controlled by the MPPT technique Open circuit voltage [VOC] 43.66 [V]
command, precisely by the duty cycle signal. In gen-
Short circuit current [ICC] 5.69 [A]
eral, this command consists to increasing or decreas-
ing the voltage produced by the PVG so that either it Number and type of the cells 72 monocristalline
converges to the MPP voltage (VMPP), by a voltage
step-down converter or voltage-boosting converter is
introduced respectively, or both the two at the same direct coupling voltage is always lower than that at the
time when the circumstances require (variations in the
MPP (VMPP). In this case, the Buck convertertype is
weather conditions and type of the load). In this article
we will work with three different resistive loads as requested because it makes it possible to reduce its
example RL = 4 ohm, RL = 10 ohm and RL = 40 ohm. output voltage (VL) (Fig. 7) to converge that of its input
Figure 4 shows the responses of these three resistors with (VPV) to the MPP voltage (VMPP) with the control of the
a direct coupling operating mode (without MPPT) with duty cycle signal generated by the MPPT technique (In
our PVG (Luxor 190M), under the climatic conditions of our case P&O). When the load is RL = 40 ohm, whatever
a solar irradiation varying between 1200 to 200 W/m2 and the weather conditions indicated, the direct coupling
an ambient temperature of 25°C. voltage is higher than that of the MPP (VMPP), hence
From Fig. 4, we can make an analysis that will the need to introduce the Boost converter type for
allow us to specify what type of DC/DC power con- player the impedance adapter role, and doing the
verter will be required for each of the selected resistive opposed role of the Buck converter. Finally, for the
loads. For the load RL = 4 ohm, we observe that what- load RL = 10 ohm, where we meet the critical point
ever the solar irradiancevariations between 1200 to (A). This point indicate us that in order to achieve a
200 W/m2 with an ambient temperature of 25°C, the good impedance matching it is imperative that, a given

(a) (b)
8 250

T = 25°C
7
PMPP
ICC 200
6 1000 W/m2

5
IMPP 150
Current, A

Power, W

100
3

2
50
1 VMPP VOC

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 3. (a) (V-I) and (b) (V-P) characteristic of the Luxor LX-190M module.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

238 BA et al.

(a) (b)
8 250
Rop (MPPT)
RL = 4 ohm Rop (MPPT) T = 25°C
7 1200 W/m2
200
6 1000 W/m2 RL = 4 ohm RL = 10 ohm
RL = 10 ohm
5
Current, A

800 W/m2 150

Power, W
4 A
600 W/m2 A
3 100
400 W/m2
2 RL = 40 ohm
RL = 40 ohm 50
200 W/m2
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 4. (a) (I-V) and (b) (P-V) characteristics of the direct coupling responses of the three resistive loads.

moment change the operating mode of the DC/DC Figure 6 shows the characteristics of our PVG
power converter. Either from buck mode to boost before and during the partial shading effect. Before the
mode or vice versa, so the point of (A) is equivalent to shading effect (with 1200 W/m2 of irradiance for
the operating modes transition moment, in this case 72 cells) the PVG must work on MPP with a power
the Buck-Boost converter type is solicited to overcome peak P = 230.5 Wc. Other hand with the shading effect
this problem. on the 10 cells (from 1200 to 400 W/m2 of irradiance), we
have two MPPs; MPP1 with a power peak P1 = 187 Wc,
The partial shading effect is a problem that can and MPP2 with a power peak P2 = 91.3 Wc (MPP1 >
affect the efficiency of PVG production by creating MPP2).
multiple MPPs, which can create a problem for MPP
Figure 7 and Table 2 show respectively, the topolo-
research (MPPT techniques). In the literature to solve gies and parameters of these three types of DC/DC
this problem, we use a Bay-pass diode (BP diode) converters. [11–14].
across of each PV cells [2].
Where : α and f are respectively the duty cycle and
In our simulation, we impose the partial shading the frequency of the opening and closing of the switch
effect condition on 10 PV cells among the 72 cells con- converter K, RPV is the PVG resistance, RL load resis-
nected in series, which constitute our PVG, with an tance, VL and VPV respectively voltage of the load and
irradiation of 1200 W/m2 towards 400 W/m2 (Fig. 5). PVG, IL and IPV respectively load and PVG current.

IPV Perturb and Observe Method (P&O)

+ This method is based on the current and voltage
instantaneous measurement, delivered by the PVG,
1200 W/m2 and then the power-voltage ratio (ΔP/ΔV) variation
62 cells stat. Once a perturbation Vref is imposed on the PVG
– VPV voltage. The sign of (ΔP/ΔV) is drawn to determine
Partial
– whether to increment the disturbance voltage (Vref) by
shading a constant value (+K) to increase the PVG voltage
effect 10 cells
(VPV) when the sign (ΔP/ΔV) > 0 or decrement Vref by
400 W/m2 a value of (–K) to decrease the PVG voltage, when
Bay-pass diode (ΔP/ΔV) < 0 for converging VPV to VMPP (Fig. 8). On
+
the other hand, if (ΔP/ΔV) = 0 hold Vref constant, and
Fig. 5. The partial shading effect on the PVG. in this case we see some ripples around the MPP, and

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

 COMPARATIVE STUDY OF DIFFERENT DC/DC 239

(a) (b)
8 250
MPP
MPP T = 25°C
7
200
6 MPP1 MPP1

5
1200 W/m2 for 72 PV cells 150
Current, A

Power, W
4

100
3
2
1200 to 400 W/m
for 10 PV cells affected MPP2
2
the partial shading effect 50
MPP2
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 6. (a) (I-V) and (b) (P-V) characteristics of the partial shading effect on the PVG.

(a) (b)
IPV L D IL IPV K L IL
+ + + +
VPV K C VL VPV D C VL

– – – –

(c)
IPV K D
+ –
VPV L C VL

– +
IL

Fig. 7. (a) Boost converter, (b) Buck converter and (c) Buck-Boost converter.

this is one of the disadvantages of using the P&O perature of 25°C. The values of the components of
method [15]. each power converter type are given in Table 3.
Figures 9 and 10 respectively show the responses of
RESULTS AND DISCUSSIONS two resistive loads; RL = 40 ohm with the Boost con-
In these simulation tests, we choose to work in spe- verter type and RL = 4 ohm with the Buck converter
cific climatic conditions; with a solar irradiance vary- type under the weather conditions indicated in Fig. 4.
ing between 1200 to 200 W/m2 and an ambient tem- For the load RL = 40 ohm, we had seen that its direct

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

240 BA et al.

Table 2. Parameters of three types of converters used coupling voltage is always higher than that of the MPP
whatever the weather conditions. For this raison, we
Parameter Boost Buck Buck-Boost introduced the Boost converter type which makes it
possible to raise the output voltage (VL) to decrease
VPV (Fig. 7a), in order to make this later converges to
VL 1 α α
VMPP with the help of signal control generated by the
VPV 1−α 1−α
MPPT (P&O) technique. For the load RL = 4 ohm
also, its direct coupling voltage is always remains lower
IL
1−α 1 1−α than that of the MPP under the weather conditions
I PV α α indicated in Fig. 4. hence we need to introduce the
Buck convertertype which makes it possible to lower
the voltage of its output (VL) to better increase its input
(1 − α) RL
2
RL voltage, which is VPV (Fig. 7b) so that the latter also
RPV (1 − α)2 RL
α² α² converges towards VMPP with the help of the MPPT
(P&O) technique.

L (1 − α) αRL
2
(1 − α ) RL (1 − α) RL
2 Figures 11 and 12 shows the responses of two types
2f 2f 2f
of converters; respectively the Boost and Buck con-
verters when the load used is RL = 10 ohm. We have
already seen that the direct coupling response curve of
C
αVL (1 − α ) VL αVL this load intersects the maximum power point line,
ΔVL RL f 8ΔVL Lf ² ΔVL RL f which is represented by point (A) in Fig. 4; this point
is the point where these two converters stop to make
the system working at the MPP. In this point of (A), we
Table 3. DC/DC power converters component values also find an overlap between the two curves. In addi-
tion, this pointcan be referred to as the Boost to Buck
Composantes Boost Buck Buck-Boost transition point or vice versa. This point is inthe [800,
600 W/m2] equivalent solar irradiance interval, which
L, H 10–5 10–5 10–4 is considered to be the critical interval of the system
when the type of converter used is either Buck or
C, F 10–3 6 × 10–3 10–3 Boost. Moreover, probably in a sunny day we can meet
twice this critical point, to overcome this problem we
f, KHz 50 50 50 introduce the Buck-Boost converter type (Fig. 7c).
See Fig. 13.

(b)
Input: V(t), I(t)

Calcul
(a) P(t) = V(t), I(t)

PMPP dP = P(t) – P(t – 1)

No Yes
dP > 0
P

dP > 0 dP > 0 No Yes No Yes

dV > 0 dV > 0

dV > 0 dV < 0
Vref = Vref + K Vref = Vref – K Vref = Vref + K

V Return

Fig. 8. (a) MPP search technique, (b) P&O algorithm.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

 COMPARATIVE STUDY OF DIFFERENT DC/DC 241

(a) (b)
8 250
RL = 40 ohm MPPT
7 MPPT

200
6

5
150
Current, A

Power, W
4

3 100

2
50
1
Direct coupling Direct coupling
0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 9. MPPT response for the load RL = 40 ohm with the Boost converter.

(a) (b)
8 250
RL = 4 ohm MPPT
7 MPPT

Direct coupling 200

6

5 Direct coupling
150
Current, A

Power, W

100
3

2
50
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 10. MPPT response for the load RL = 4 ohm with the Buck converter.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

242 BA et al.

(a) (b)
8 250
RL = 10 ohm MPPT
7 MPPT
200
6

5
150
Current, A

Power, W
4
A A
100
3 Direct coupling
Direct coupling
2
50
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 11. MPPT response for the load RL = 10 ohm with the Boost converter type.

(a) (b)
8 250
RL = 10 Ohm
7
200 MPPT
6

5 MPPT
150
Current, A

Power, W

4 A A
100
3 Direct coupling
Direct coupling
2
50
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 12. MPPT response for the load RL = 10 ohm with the Buck converter type.

Table 4, Figs. 14 and 15 show the importance of the CONCLUSIONS

use of a Bay-pass diode in the partial shading effect The study of the voltage-current and voltage-power
conditions case, we find that whatever the type of characteristics of a PVG with the direct coupling loads
DC/DC power converter and the load used, the system responses as a function of the weather conditions in a
cannot work on the MPP1 without the Bay-pass diode. precise area, are the important steps to arrive at the

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

 COMPARATIVE STUDY OF DIFFERENT DC/DC 243

(a) (b)
8 250
RL = 10 ohm MPPT
7 MPPT
200
6

5
150
Current, A

Power, W
4

100
3
Direct coupling
Direct coupling
2
50
1

0 10 20 30 40 50 0 10 20 30 40 50
Voltage, V Voltage, V

Fig. 13. MPPT response for the load RL = 10 ohm with the Buck-Boost converter type.

build of a perfect impedance matching to better opti- (MPPT), just by comparing the direct coupling volt-
mize the PV solar system productivity. ages responses to the optimal voltages (MPP).
We have taken as an example to work with three dif- In this case, when we decide to work under specific
ferent types of loads, which represent the three possi- weather conditions (taking in to account the partial
ble cases that we can see in the process of the stand- shading effect conditions). If the direct coupling volt-
ages are lower than those of the MPP, we use the Buck
alone PV solar systems optimization. In addition, converter type. On the other hand, if the direct cou-
depending on their direct coupling responses (without pling voltages is higher than the MPP, the Boost con-
the MPPT technique using), we were able to deter- verter is used. Under some conditions, we encounter
mine which types of DC/DC power converter to loads whose direct coupling responses force us to
introduce in order for the system to work optimally decrease the voltage at the given moment and increase

Table 4. DC/DC power converters performance in the partial shading effect condition

Type of DC/DC Power converter with the load

Shading
Buck-Boost
effect Buck (RL = 4 ohm) Boost (RL = 40 ohm)
(RL= 4, 10 and 40 ohm)
condition

with BP diode without BP diode with BP diode without BP diode with BP diode without BP diode

Before MPP MPP MPP MPP MPP MPP

During MPP1 MPP2 MPP1 MPP2 MPP1 MPP2

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

244 BA et al.

(a)
1400

Irradiance, W/m2
1200
1000 62 cells without the partial shading
72 cells without the partial effect (1200 W/m2 and 25°C)
800 shading effect
600 (1200 W/m2 and 25°C) 10 cells in the partial shading
400 effect (400 W/m2 and 25°C)
200
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Time, s

(b)
300
With a Bay-pass diode (MPP1)
MPPT, W

200

100
Without a Bay-pass diode (MPP2)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Time, s

Fig. 14. (a) Irradiation, (b) Buck-Boost performance with RL = 10 ohm in partial shading effect.

(a)
300
With a Bay-pass diode (MPP1)
MPPT, W

200

100
Without a Bay-pass diode (MPP2)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Time, s

(b)
300
With a Bay-pass diode (MPP1)
MPPT, W

200

100
Without a Bay-pass diode (MPP2)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Time, s

Fig. 15. (a) Boost and, (b) Buck performance with respectively RL = 40 and 4 ohm in partial shading effect.

at another moment; hence, we are obliged to use the REFERENCES

Buck-Boost converter type to overcome these kinds of
1. Luo, F.L. and Ye, H., Advanced DC/DC Converters,
problems.
Boca Raton: CRC, 2004.
In conclusion, the Buck-Boost converter and the 2. Mohapatra, A., Nayak, B., Das P., et al., A review on
Bay-pass diode for each photovoltaic cell are two nec- MPPT techniques of PV system under partial shading
essary elements to have a more efficient and better condition, Renewable Sustainable Energy Rev., 2017,
robust MPPT optimization PV system. vol. 80, pp. 854–867.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018

 COMPARATIVE STUDY OF DIFFERENT DC/DC 245

3. Enrique, J.M., Duran, E., Sidrach-de-Cardona, M., 10. Saravanan, S. and Ramesh Babu, N., RBFN based
et al., Theoretical assessment of the maximum power MPPT algorithm for PV system with high step up con-
point tracking efficiency of photovoltaic facilities with verter, Energy Convers. Manage., 2016, vol. 122,
different converter topologies, Sol. Energy, 2007, pp. 239–251.
vol. 81, no. 1, pp. 31–38.
4. Basoglu, M.E. and Cakir, B., Comparisons of MPPT 11. Kolsi, S., Samet, H., and Amar, M.B., Design analysis
performances of isolated and non-isolated DC/DC of DC-DC converters connected to a photovoltaic gen-
converters by using a new approach, Renewable Sus- erator and controlled by MPPT for optimal energy
tainable Energy Rev., 2016, vol. 60, pp. 1100–1113. transfer throughout a clear day, J. Power Energy Eng.,
2014, vol. 2, no. 1, pp. 27–34.
5. Bellia, H., Youcef, R., and Fatima, M., A detailed
modeling of photovoltaic module using MATLAB, 12. Fan, H., Design tips for an efficient non-inverting
NRIAG J. Astron. Geophys., 2014, vol. 3, no. 1, pp. 53– buck-boost converter, Analog Appl. J., 2014, pp. 20–25
61.
6. Ahmed, J. and Salam, Z., An improved perturb and 13. Glasner, I. and Appelbaum, J., Advantage of boost vs.
observe (P&O) maximum power point tracking buck topology for maximum power point tracker in
(MPPT) algorithm for higher efficiency, Appl. Energy, photovoltaic systems, Proceedings of 19th Convention of
2015, vol. 150, pp. 97–108. Electrical and Electronics Engineers in Israel, 1996,
7. Mukti, R.J. and Islam, A., Modeling and performance pp. 355–358.
analysis of PV module with maximum power point
14. Farahat, M.A., Metwally, H.M.B., and Abd-Elfatah, M.A.,
tracking in Matlab/Simulink, Appl. Sol. Energy, 2015,
vol. 51, no. 4, pp. 245–252. Optimal choice and design of different topologies of
DC/DC converter used in PV systems, at different cli-
8. Radjai, T., Rahmani, L., Mekhilef, S., et al., Imple- matic conditions in Egypt, Renew. Energy, 2012,
mentation of a modified incremental conductance vol. 43, pp. 393–402.
MPPT algorithm with direct control based on a fuzzy
duty cycle change estimator using dSPACE, Sol. 15. Mohanty, P., Bhuvaneswari, G., Balasubramanian, R.,
Energy, 2014, vol. 110, pp. 325–337. et al., MATLAB based modeling to study the perfor-
9. Bendib, B., Krim, F., Belmili, H., et al., Advanced mance of different MPPT techniques used for solar PV
fuzzy MPPT controller for a stand-alone PV system, system under various operating conditions, Renewable
Energy Proc., 2014, vol. 50, pp. 383–392. Sustainable Energy Rev., 2014, vol. 38, pp. 581–593.

APPLIED SOLAR ENERGY Vol. 54 No. 4 2018