IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,

Bhubaneswar, India

Extraction of Maximum Power from a Solar

PV System using Fuzzy Controller Based
MPPT Technique
SaritaSamal Prasanta Kumar Barik
School of Electrical Engineering Department of Mechanical and Electrical Engineering,
KIIT University, Bhubaneswar,India CAET, OUAT, Bhubaneswar, India
saritaruchy@gmail.com prasantbarik05@gmail.com

Sandeep Kumar Sahu

School of Electrical Engineering
KIIT University, Bhubaneswar,India
sandeep.sahu.1996@gmail.com

Abstract— Two types of MPPT techniques are was confined to PV module characteristics
focused in this paper i.e. the conventional Perturb simulation in all the above. The mathematical
and Observation method and fuzzy logic controller. It expressions determining the PV module (as well as
is seen that the tracking speed of fuzzy logic PV cell) are also represented. For each expression,
controller is consistent and improved in comparison Simulink model is represented with numerical
to traditional P&O controller. However, the results for constant irradiation values (1000W/m2)
disadvantages of P&O controller is that it has very and temperature. When building a new
poor dynamic response and the operating point keeps photovoltaic power system it is very important to
fluctuating around the maximum power point during
consider maximum power point tracking (MPPT)
steady state operation. So in this paper the emphasis
is given to the fuzzy logic controller design and a as it is required for extraction of maximum power
comparison is made between fuzzy logic controller output from a PV array under varying atmospheric
and P&O controller. The algorithm implemented conditions for maximum power output. Many
here was used to track the maximum power of a day. researchers and industry delegates from all over the
The implementation used a fuzzy logic controller and worldhave developed several MPPT algorithms.
a DC-DC converter is used to maintain the PV output Some of these algorithm like perturbation and
power at the highest level all the time. All the results observation(P&O) method, fuzzy logic control
are simulated using MATLAB/SIMULINK. method, linear approximation method, incremental
Keywords—Boost Converter, Maximum power point conductance method, voltage feedback method, hill
tracking, Perturbation and observation, PV Modelling climbing method, actual measurement method and
so on [14-15]. Appropriate MPPT method
I. INTRODUCTION alongwith good weather conditions are required for
The application of photovoltaic (PV) has achieved implementing maximum performance of a
an exponential rise for past two decades from off- photovoltaic system [9-11]. This paper mainly
grid to grid connected PV systems. The electric focuses on studying and comparing execution
energy produced by the PV array can be utilized in efficiency, advantages, disadvantages for two
the best way by delivering it directly to utility grid, power-feedback type MPPT methods, including
without using storage system (battery banks) [1]. perturbation & observation (P&O) and fuzzy logic
The performance analysis of newly developed (FL) methods. For implementation of modelling
systems requires mathematical functional models and simulations tasks, and to compare execution
for PV module research. Field professionals do not efficiency and accuracy for selected MPPT
readily adopt these developed systems for methods Matlab/Simulink is used in this paper.
minimising failure rate. Therefore, it requires II. DESIGNING OF PV MODULE
simplified Simulink modelling of PV module for
analysis purpose. In the literature [2–11] basic A. Equivalent Circuit
structure of single diode PV system have been Figure 1 represents a single diode model which is
represented. For adjusting the I-V curve by using widely used as compared to other PV module
artificial intelligence [12, 13] some authors have design. In this circuit, Rsh is shunt resistance, I is
put forward some indirect methods. Although open circuit currentof a solar cell, Io is diode
interesting but these methods are complex, saturation current, Iph is the light-generated current
inapplicable and needs more calculation. Modelling which depends upon solar radiation and cell
temperature, Ish is shunt resistance current which

978-1-5386-4581-9/18/$31.00 ©2018 IEEE

 IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,
Bhubaneswar, India

flows through Rsh and flows between the n and p To get the I-V and P-V characteristics of Ipv model
layers, Rse is series resistance which represents the under varying irradiation with constant temperature
losses due to flowing current across highly resistive and constant irradiation with varying temperature,
emitter and contacts, Voc is terminal voltage of a simulation steps have been discussed in details.
solar cell, respectively.
I

Rse
Ish
Iph Voc
Io Rsh

Fig. 1 Solar cell equivalent circuit

In this circuit the mathematical expression for cell
current in single diode model is obtained by
applying KCL,
I= Iph-Io-Ish(1)
Where,
Iph= [Isc+Ki (Tk-T)] *G/1000 (2)
Iph is photocurrent under standard test conditions
(STC), with reference solar radiation of 1000 W / Fig 2. Simulation Module output current Ipv
m2 at solar spectrum of 1.5A and reference
temperature of solar cell Tk of 25°C. T is III. PV MAXIMUM POWER EXTRACTION
instantaneous solar cell temperature, Ki is current SYSTEM DESIGN
temperature coefficient and G is an instantaneous The PV module output power changes significantly
solar radiation. as there is variation in irradiation and temperature.
To extract the maximum power of from solar
B. Module Reverse Saturation Current (Irs) photovoltaic module and transfer this power to the
Module reverse saturation current, Irs, is expressed load, the maximum power point (MPPT) algorithm
as, is used [15]. To transfer the maximum power of the
Irs= Isc/ [exp (q*Voc/Ns*k*A*T)-1] (3) PV module to the load, a DC-DC converter is used
Where q is the electron charge (1.6 × 10í19 C), Voc as shown in Figure 14, which acts as an interface
is Solar module open-circuit voltage (21.24V), Ns is between the load and the module. Here, step
the number of cells connected in series (36), A is up/step down type DC-DC Converter is used. The
the ideality factor (A=1.6). and k is the Boltzmann load impedance as seen from source side is
constant (k= 1.3805 × 10í23 J/K). changed by varying the duty cycle of PWM
control signal and hence it coincides with the
C. Module Saturation Current (I0) maximum power point of the source to transfer the
Variation of module saturation current I0takes place maximum power.
with respect to cell temperature. It is expressed as, PV MODULE DC-DC LOAD
I0=Irs [T/Tr] 3*exp [q*Eg0/Ak*{(1/Tr)-(1/T)}] (4) VPV IPV

Where Eg0 is the band gap energy of semiconductor

(For polycrystalline Si at 25ƕC, Eg0=1.1 eV).
V

Simulation of equation (4) has been done and DIFF. MPPT PWM
represented in Figure 4. Here, the inputs are TECHNIQUES CONTROL

module reverse saturation current, module

operating temperature and reference temperature.
Fig.3 Block diagram of the DC-DC converter for MPP operation
A. Power Electronic Circuit
D. Module Output Current (IPV)
To obtain the maximum performance of the
The PV module output current of single diode
maximum power point, PV modules are always
model isIPV represented in Figure 1 is described by
used with DC-to-DC converters. Buck, boost, and
a basic equation and is expressed as,
buck-boost are types of converters used for this
Ipv=Np*Iph-Np*I0 [exp {q*(Vpv+Ipv*Rse)*Ns*A*k*T-
purpose. For grid-connected applications boost
1] (5)
converter is used where as buck-boost
configuration is used for battery charging
E. Simulink Model of IPV
applications. Figure 4. represents the configuration
Inputs for this model are insolation, VPV is
of boost converter and it is composed of load
Varied from 0 to 21.5V. With the help of setup
resistance R, filter capacitor C, diode D, control
shown in Fig. 5, simulink model of IPV is simulated.
 IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,
Bhubaneswar, India

switch S, boost inductor L, and DC input voltage

source Vs. L D

The voltage gain of boost converter when switch

operated with a duty ratio D is expressed as,
Mv= Vo/Vs= 1/1 – D (6) VS
S
C R Vo

Where Vo is output voltage, Vs is input voltage, and

D is duty cycle of PWM (pulse width modulation)
signal and used to control ON and OFF states of
MOSFET.
Table 1. Electrical Parameters of 250W Panel Figure 4: DC to DC boost converter configuration
Description Rating
Swtiching frequency 20 kHz
Power diode 1N5408 D. MPPT Methods
MOSFET IRF P460 For maximum power point tracking many methods
Capacitor 330 ȝF are used. Some of the methods are:
Resistive Load 50ȍ, 50W • Constant Current method
Inductor 120 ȝH
• Constant Voltage method
TABLE I. • Parasitic Capacitance method
DC-TO-DC BOOST CONVERTER • Incremental Conductance method
COMPONENT VALUES • Perturb and Observe method
For value of inductance L >Lb, the operation of boost
converter is in continuous conduction mode where, IV DESIGN OF (P&O) MPPT
Lb= (1 í D2) DR/ 2 f (7)
Where Lb is smallest inductance value required for Start

continuous conduction. As current supplied to

output RC circuit is discontinuous. Therefore, to Read V(k) & I(k) from panel

limit the output ripple voltage a large filter & calculate P(k)=V(k)*I(k)

capacitor is required. Cmin is minimum value of

filter capacitor. It supplies output DC current to the Delay P(k) & V(k) by k-1 instant P(k-1),
V(k-1)

load when diode D is off. The smallest value of

filter capacitance results in ripple voltage Vc and it
is expressed as, οሺሻൌሺሻǦሺǦͳሻ
οሺሻൌሺሻǦሺǦͳሻ

Cmin = DVo/ VrRF(8)

Table 1 gives the DC-DC boost converter design
¨P >0
component values which are used for simulation.

C. Design of MPPT ¨V < 0 V<0

Figure 7 shows the DC-DC converter (with

component values in Table.1andconfiguration
given in Figure 4) simulation with battery supply D = D + ¨D D = D - ¨D D = D - ¨D D = D + ¨D

.The solar cell has very low efficiency. Therefore,

to boost the efficiency some methods are To Switch

implemented to balance the source and load

properly. Maximum Power Point Tracking (MPPT) Fig.5 MPPT flowchart
method is one of the methods which are used to
obtain the maximum possible power from a
variable source. It is difficult to use photovoltaic
systems to power certain load because the I-V
curve in photovoltaic systems is non-linear. So
boost converter is used and its duty cycle is varied
by using a MPPT algorithm. By the help of
simulation with developed circuit mode lthe data is
collected for the design of MPPT. The results are
represented in Table 2.

Fig.6 Simulink model of P&O algorithm

 IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,
Bhubaneswar, India

E. Perturb and Observe method So fuzzy control algorithm have the ability to
It is widely used method. Minimum sensors are improve the tracking performance for both linear
used in this method. In this method, sampling of and nonlinear loads as compared with the classical
operating voltage is done and operating voltage is methods. As fuzzy logic does not use complex
changed in a specific direction by using algorithm mathematical equation therefore it is also
and therefore it samples ݀ܲ / ܸ݀. The algorithm appropriate for nonlinear control. Figure.10
increases the voltage value towards MPP until ݀ܲ / represents the block diagram of fuzzy logic
ܸ݀ is negative if ݀ܲ / ܸ݀is positive. This iteration controller (FLC). The shape of membership
continues until the algorithm arrives at MPP. When functions of the rule base is one of the factor on
there is a large variation in solar irradiation then which the behaviour of a FLC depends
this algorithm is not suitable. The voltage perturbs
around the maximum power point (MPP) and never START
actually reaches an exact value.
A. Simulation of P&O MPPT
Initialize P(k-1) = 0

Measure V(k) , I(k)

Then find P(k)

Find ¨P & ¨I

Fuzzification

Fuzzy
Fig.7 Simulation of P&O MPPT Rule Base Inference Set

B. Result
Defuzzification

D (Duty Cycle)

Fig.9 Flowchart of the proposed FLC method

Membership functions and rule base are designed
with the help of Fuzzy Logic Tool Box in
MATLAB. The membership function graphical
view for error is represented byFig.12 .Change of
error is represented byFig.13 and fuzzy logic
controller duty cycle is represented in Fig.14.
Knowledge Base

E(k)
Decision
Inputs Making Output
Fuzzy logic is a dynamic control method. It is CE(k) Logic

recognized by multivariable consideration and Defuzzification

Interface
multi-rules-based resolution. Over the last decade Fuzzification
Interface
Inference
Engine
Fuzzy MPPT is very much popular. Working with
inaccurate inputs, the ability to handle non-linearity Fig. 10 Fuzzy logic block diagram
and not having a precise mathematical model are
some of the advantages of fuzzy logic controllers.
The Fuzzy MPPT flow chart is represented in Fig.9
and proposed Fuzzy MPPT Simulink model is
represented in Fig.15.It consists of two inputs and
one output. Error (E) and change of error (CE) are
two FLC input variables. Duty cycle (D) is output
variable.
 IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,
Bhubaneswar, India

Fig.14 Fuzzy logic output (D)

Fig. 11 Fuzzy logic Implementation in Simulation
Graphic view of the membership function for (a)
error signal (b) change of error signal and (c) duty
cycle. Different number of subsets has been used
for rule settings of fuzzy logic MPPT. In this case,
seven subsets based on forty- nine rules were used.
The tuning of forty nine rules represents a better
precision and dynamic response but it is time
consuming.

Fig.12 Fuzzy logic input Error (E)

Fig.15 Fuzzy logic rule view

V. RESULTS OF SIMULATION AND

DISCUSSION
The proposed model of Fuzzy MPPT based solar
PV system is realized in the MATLAB SIMULINK
environment which is represented in Fig. 9 .Here a
comparison has been made with traditional P&O
technique and fuzzy MPPT technique under the
operating conditions, accepting constant
temperature and an isolation of 1000 W/m2. In
Fig.13 Fuzzy logic input change of Error (CE)
P&O technique, the incremental current step size
uses the appropriate value that is processed by trial
and error technique. Current control and FLC are
two important parts of MPPT control which is
represented in Fig. 9.
 IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28-30, 2018,
Bhubaneswar, India

[6] Mohamed Azab, “A New Maximum Power

Point Tracking for Photovoltaic Systems”,
World Academy of Science, Engineering and
Technology, Vol. 34, October 2008.
[7] Larbes, S.M. AÕ¨t Cheikh*, T. Obeidi, A.
Zerguerras, “Genetic algorithms optimized
fuzzy logic control for the maximum power
point tracking in photovotaic system", Science
Direct Renewable Energy 34, January 2009,
pp.2093-2100.
Fig. 15 Simulink model of Fuzzy logic MPPT [8] Cheikh M. S. A., Larbes C., Kebir G. F. T. and
ZerguerrasA.,“Maximum power point tracking
using a fuzzy logic control scheme", Revue des
Energies Renouvelables,Vol. 10, No. 32,
September 2007, pp. 387 – 395.
[9] Sofai. Lalouni, Djamila. Rekioua, “Modeling
and Simulation of Photovoltaic System using
Fuzzy Logic Controller", IEEE International
Conference on Developments in Systems
Engineering, 2009.
[10] S. Sarita, and P. K. Hota. "Design and analysis
of solar PV-fuel cell and wind energy based
microgrid system for power quality
improvement." Cogent Engineering 4, no. 1
pp.1402453, 2017.
[11] S. Samal, and P.K, Hota, “Power Quality
Improvement by Solar Photo-voltaic/Wind
Energy Integrated System Using Unified Power
Quality Conditioner”, International Journal of
Power Electronics and Drive Systems, 8(3),
p.1424, 2017.
[12] Christopher A. Otieno, George N. Nyakoe,
Cyrus W. Wekesa, “A Neural Fuzzy Based
Maximum Power Point Tracker for a
Photovoltaic System”, IEEE Africon,
Fig.16 Result of Fuzzy MPPT September 2009.
[13] S. SAMAL, and P.K., HOTA, “Power Quality
TABLE II. COMPARISON BETWEEN DIFFERENT MPPT Improvement by Solar Photo-voltaic/Fuel Cell
SYSTEMS Integrated System Using Unified Power Quality
Simulation run time 8 msec Conditioner”, International Journal of
Different Vin Iin Pin Vo/p Io/p Po/p Renewable Energy Research (IJRER), 7(4),
MPPT pp.2075-2084. , 2017
system [14] N. Pandiarajan and R. Muthu, “Mathematical
P&O 12.4 2.55 30.8 29.8 0.65 17.88 Modeling of Photovoltaic Module with
FUZZY 12.4 2.55 30.8 24.2 0.51 12.35 Simulink,” in Proceedings of the
InternationalConference on Electrical Energy
IV. CONCLUSION Systems(ICEES’11), Jan 2011
The proposed method of MPPT is faster in [15] S.Samal, and S. K. Das. "Solar Energy Fed to
3-Phase Induction Motor using Matlab
comparison to conventional P&O method when
Simulink and their analysis." Materials Today:
parameters like output voltage, output current and Proceedings 4, no. 14,pp.12615-12624,2017
efficiency are compared. From Table. II it is clear
that the system performance can be improved by
proposed MPPT technique using fuzzy logic. The
proposed technique can be implemented in the real
PV system for future work.

References
[1] N. Pandiarajan and R. Moth, “Viability analysis on
photovoltaic configurations", Proceedings of the
IEEE Region 10 Conference (TENCON ’08),
Hyderabad, India, November 2008.
[2] “PV Balance of Systems Conference Berlin,
Germany,” June 2011, http://www.PV insider.com/.
[3] Chao Zhang, Dean Zhao, “MPPT with Asymmetric
Fuzzy Control for Photovoltaic System”, IEEE
Africon, 2009.
[4] Christopher A. Otieno, George N. Nyakoe, Cyrus W.
Wekesa,“A Neural Fuzzy Based
Maximum Power Point Tracker for a Photovoltaic
System”, IEEE African, September 2009.
[5] Neural network in maximum power point tracker for
PV systems”, Science Direct Electric Power Systems
Research, July 2010, pp.43–50.