Transactions Template


 

  

JOURNAL OF ENGINEERING RESEARCH AND TECHNOLOGY, VOLUME 7, ISSUE 2, OCTOBER 2020 

  

1  

 

Maximum Power Point Tracking Control for Grid-Connected 

Photovoltaic System under Partial Shading Conditions 

Mohammed S. Ibbini and Areen G. Al-Obeidallah 

DOI: https://doi.org/10.33976/JERT.7.2/2020/1 
 

Abstract—This manuscript uses an active bypass circuit to preserve the photovoltaic (PV) modules from a 
significant reduction in power generation, which is transferring from partial shading. The active bypass circuit 
controls each PV module separately, including the shaded ones, by detecting the shaded region on these modules 
using an image processing technique. The active bypass circuit ensures that every PV module on the solar system 
is working at its maximum power operating point. Under partial shading conditions, the conventional maximum 
power point tracking (MPPT) methods have usually failed to track the global MPP. In order to overcome this 
limitation, this paper attempts to combine a fuzzy logic technique and an active bypass circuit to ensure the 
detection of a unique global power point (GMPP). A current source inverter (CSI) is used with a double-tuned 
resonant filter to eliminate the undesirable harmonics on the direct current (DC) side. In addition, an LC filter is 
used in the AC side to reduce the switching harmonics. The simulation results demonstrate improvement in the 
performance of the proposed MPPT method under partial shading conditions. Hence, the combination of the active 
bypass circuit and the image processing technique, based on a fuzzy logic decision-making approach, provides an 
effective process for the detection of the shaded region on the surface of the PV module. In addition, this 
combination ensures a unique global maximum power point (GMPP)for the proposed PV solar system. 

Index Terms— Photovoltaic (PV) Module, Image Processing, Active by Bass Circuit, Shadow Detection, Fuzzy 
Logic, Maximum Power Point Tracking (MPPT), Grid-Connected.  

 

I. INTRODUCTION

The performance of a photovoltaic (PV) module mainly 

depends on the amount of solar irradiance that can be ab-

sorbed by each module. PV modules must commonly be con-

nected in series in order to generate the desired power. If any 

part of these modules is prohibited from receiving light by 

means of partial shading, their power generation will be re-

duced. As a result of partial shading on a PV module, a sig-

nificant decrease in the total PV power generation is ob-

served. To solve this problem, conventional approaches com-

monly connect a bypass diode in parallel with each PV mod-

ule [1- 4]. However, the shaded PV modules cannot produce 

their inherent power and the overall power generation solar 

system is still smaller than the power generation without 

shading conditions [5]. 

In order to handle the problem of partial shading on PV 

modules, this manuscript uses an active bypass circuit, a gen-

eration control circuit (GCC) [6-7]. In fact, the active bypass 

circuit controls each PV module separately, including the 

shaded ones, by detecting the shadow region on these mod-

ules based on the image processing technique [8-9]. The ac-

tive bypass circuit ensures that every PV module of the solar 

system is working at its maximum power operating point [10-

11]. To track the maximum power point of the proposed solar 

system, a new method, based on a combination of an active 

bypass circuit and a fuzzy logic technique, is proposed. This 

manuscript is organized as follows: In Section 2, the proposed 

PV solar system is presented. Section 3 presents the shadow 

detection method on a PV array using a digital image pro-

cessing technique. Section 4, explores the proposed configu-

ration of the active bypass circuit and its control principle. 

Section 5 presents the fuzzy logic control technique for track-

ing the maximum power point. Section 6, presents the simu-

lation results. Finally, in Section 7 conclusions are given and 

discussion.   

II. PROPOSED PV SOLAR SYSTEM 
DESCRIPTION 

Fig.1 illustrates the circuit configuration for the proposed PV 

solar system, where two PV modules consist of ten solar cells 

are connected in series in order to generate the required volt-

age and circuit. monitoring camera gives a quick indication 

of an existing shadow on these modules by using the proposed 

digital image processing technique. The digital processing 

technique is based on a fuzzy logic decision, making system 

permitting the determination of the shadow type. The pro-

posed technique enables the active bypass circuits, which are 

connected in parallel with each PV module, to generate a con-

trol signal from a pulse width modulator (PWM) to ensure the 

detection of unique MPP for solar system. A current source 

inverter (CSI) with a double tuned parallel resonant circuit is 

connected to the DC side of the solar system to boost the DC 

input voltage with very reduced value of unwanted harmon-

ics. The AC side of the system is connected to the grid using 

https://doi.org/10.33976/JERT.7.2/2020/1


M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

2  

an LC filter to avoid switching harmonics on the AC side. 

III. PROPOSED METHOD OF SHADOW 
DETECTION ON A PV ARRAY USING 
DIGITAL IMAGE PROCESSING 

 
The need for an effective process, to detect the shadow on 

PV panels, becomes a necessity especially in a massive PV 

solar system where the shadow can cover a large part of panel. 

The main motive for using digital image processing technique 

is to establish a dynamic and effective way to detect the 

shadow on PV panels and to determine the its type [8-9]. The 

first step depends on installing a camera system to continu-

ously monitor the PV arrays. The image obtained by the mon-

itoring camera is then treated by a digital processing tech-

nique, based on a fuzzy logic decision- making system, and 

the shadow type is determined.Fig.2 illustrates the different 

stages of the overall system which will be discussed in details.  

 
A. Background of the Subtraction Method 

Background subtraction consists of extracting a fore-

ground mask from the preserved background model using the 

stationary camera. This technique is widely used for detecting 

objects or tracking targets by subtracting the input from the 

background model, which also known as the reference frame. 

The background subtraction method can be represented by the 

mathematical formula [12]: 

 

𝑀𝑡(𝑥, 𝑦)  = {
1,   if 𝐷( 𝐼(𝑥, 𝑦, 𝑡), 𝑅(𝑥, 𝑦, 𝑡))  >   𝑇              (1)

0,   otherwise 
  

 

Where: 

Mt (x, y) is the motion mask at time t, x and y are the pixel 

location variables, D is the distance between the input frame 

I (x, y, t) and the reference frame R(x, y, t), and T is the thresh-

old value, which is an empirically selected value.  

The distance between the input frame and the reference frame 

is compared with the threshold value T, if the distance value 

is geater than that of the threshold, then this indicates a differ-

ence in pixels between foreground and background which 

means the presence of a new object in the image. This dis-

tance can be presented by the following formulas: 

      
D0 =   |𝐼(𝑥, 𝑦, 𝑡)  − 𝑅(𝑥, 𝑦, 𝑡)|                                               (2)                                                                                                    
D1 = |𝐼

𝑅 (𝑥, 𝑦, 𝑡) − 𝑅𝑅 (𝑥, 𝑦, 𝑡)| + |𝐼𝐺 (𝑥, 𝑦, 𝑡) −
𝑅𝐺 (𝑥, 𝑦, 𝑡)| + |𝐼𝐵 (𝑥, 𝑦, 𝑡) − 𝑅𝐵 (𝑥, 𝑦, 𝑡)|                                                                      
 
Where D0 is the distance for the gray image, and D1 is the 

distance for the RGB image. 

The flowchart of the background subtraction method is illus-

trated in Fig.2. The main steps of this method can be summa-

rized as follow: 

1. Estimate the reference frame R(x, y, t) at time t. 
2. Subtract the reference frame from the input frame 

I(x, y, t). 

 

Apply a threshold T to the absolute difference to get the 

foreground mask. 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 2. Flowchart of the background subtraction method. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig.1. Proposed Grid-Connected PV Solar System. 

 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

3  

     The background subtraction method is sensitive to noise 

and illumination changes and it depends on the threshold 

value T, which is empirically selected value and is not a func-

tion of time similar to the input and reference frame. In order 

to improve the results of background subtraction, the disser-

tation uses a Blob detection method that highlights the new 

object in the reference image, based on the hypothesis that a 

new object has different properties, such as color distribution 

and brightness, as opposed to the surrounding regions. 

     The Blob detection method returns a zero-pixel value if 

both inputs are equal and one-pixel value if they are different. 

The detection method uses a median filter to remove any un-

expected noise from the gray image. Finally, it traces the re-

gion boundaries in a binary image. The input and the refer-

ence images of the proposed PV solar panels and the results 

of applying background subtraction and Blob detection meth-

ods are shown in Fig.3. 

A. Edge Detection Method 
Edge detection is one of the essential steps in digital image 

processing. It depends on finding the boundaries of objects 

within images by detecting the sudden changes of intermis-

sion in the brightness of an image. There are many common 

edge detection algorithms such as Roberts, Sobel, Prewitt, 

and Canny [13-18].  

Before applying the edge detection method, this dissertation 

uses a morphological process to reduce the unexpected noise 

of the image obtained from background subtraction by apply-

ing dilation and erosion operations. Dilation adds pixels to the 

boundaries of the shaded area in a PV image, while erosion 

removes pixels from these boundaries. The morphological 

process depends on the shape and size of the construction el-

ement used to treat the image [19]. Then, Canny algorithm 

will be applied to detect the shadow boundaries of a PV solar 

panel [17]. The Canny edge detector is an image processing 

technique used to detect edges in an image with the minimum 

effect of noise and more likely to detect the weak edges [20]. 

It consists of a multistage edge detection algorithm; the main 

steps for this algorithm are shown in Fig.4.  The main goal of 

using the Canny algorithm is to reduce the amount of data to 

be treated in the digital image and preserves the only signifi-

cant structural properties of the shaded area of the PV image. 

The proposed edge detection method is shown in Fig.5. 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

A. Fuzzy Logic Method 
Fuzzy logic is one of the artificial intelligent approaches that 

are based on human reasoning rather than the Boolean logic. 

This approach treats the levels of possibilities of input to ob-

tain the obvious output. In order to benefit from the flexibility 

of fuzzy logic approach, the manuscript uses this method as a 

decision-making process to classify the type of shadow on the 

PV solar panel. The proposed fuzzy logic method takes the 

values of brightness (B) and color distortion (CD) of the 

shaded area on the PV solar panel as an input parameter and 

the type of this shaded area as an output parameters. Bright-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4. A flowchart of Canny Edge Detector. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 5. The proposed Edge Detection Method. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 3. The proposed detection methods for a PV solar panel. 

 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

4  

ness is a relative expression of the intensity of the output en-

ergy of a visible light source. It can be represented as the total 

energy value, or as the amplitude where is the greatest value 

of intensity occurs at the wavelength of the visible light. In 

other words, brightness represents the intensity value of every 

pixel in the digital image. In the RGB color space, the ampli-

tudes of red, green, and blue for a specific color can tack a 

range from 0 to 100 percent of full brightness. These level 

values are represented by the range of decimal numbers from 

0 to 255 or hexadecimal numbers from 00 to FF. When the 

brightness value is 0, then the color space brightness is com-

pletely black. However, with the increase in the value, the 

brightness of the color space increases and shows different 

colors. In this manuscript, the normalized value of R, G, and 

B is used in the range [0, 1]. The following equation is used 

to calculate the value of brightness of the image by finding 

the weighted mean of intensity value for all pixels of the 

shaded are. Some data points count more strongly than others, 

in that they are given more weight in the calculation: 

 

 Bshaded= 
∑  (0.3𝑅𝑖 +0.3𝐺𝑖 +0.4𝐵𝑖)                  𝑚𝑖=1

𝑚
              (3)                                             

 

where Bshaded is the brightness value of the shaded region, and 

m is the number of pixels in the RGB color space.  

Color distortion is a deflection from rectilinear dropping of 

pixels in an image; it is a type of optical deviation. Color dis-

tortion can be calculated by using Mahalanobis distance be-

tween the arithmetic mean of red, green and blue pixel values 

of the shaded area and the arithmetic mean of pixel values of 

the reference image for PV solar panel. Mahalanobis distance 

gives a useful method of measuring how similar some set of 

pixels is to an ideal set of pixels. The following formula is 

used to calculate the value of color distortion for the shaded 

region: 

 

               𝐶𝐷𝑠ℎ𝑎𝑑𝑒𝑑 
=   𝑀𝑎ℎ𝑎_𝐷𝑖𝑠 ( 𝑀𝑒𝑎𝑛 (𝑅𝐺𝐵𝑠ℎ𝑎𝑑𝑒𝑑 , 𝑅𝑒𝑓 )                         (4) 
 

where CDshaded is the color distortion of the shaded region, RGB-
shaded is the red, green and blue pixel value of the shaded area, and 

Ref is the pixel value for the reference image.  

Mahalanobis distance is based on arithmetic mean of RGB of the 

shaded region and the variance of the expectation variables, with 

tacks the covariance matrix of all the variables in consideration. 

The region of constant Mahalanobis distance around the mean 

forms an ellipse in 2D space when only two variables are meas-

ured, or an ellipsoid or hyper ellipsoid when more variables are 

used. This distance is zero if the pixel of RGBshaded at the mean of 

the reference image, and it increases as the pixel moves away 

from the mean. The Mahalanobis distance is given by the follow-
ing equation: 

Maha_Dis   = [(μshaded – μR ) S
-1

 (μshaded – μR )
T
 ]

1/2
    (5)                                  

Where μshaded is the arithmetic mean of the shaded region, μR 
is the arithmetic mean of the reference image, and S is the covar-

iance matrix. This matrix is calculated by using the standard de-

viations of pixel values for the shaded region. 

The input membership functions of brightness and color dis-

tortion and the output membership function of the proposed 

fuzzy logic method are represented in Fig.6. The fuzzy logic 

rules are shown in Table 1.  

Bshaded is the brightness value of the shaded region, CDshaded is 

the color distortion of the shaded region, s is small, m is me-

dium, and b is big. 

 
To ensure the effectiveness of applying a digital image pro-

cessing technique in shadow detection for PV solar system 

and classifies the type of this shadow, two different images 

were obtained from a mobile camera. The results of using the 

proposed digital image processing method based on Matlab 

are shown in Fig. 7 (a) and (b). In the first case, a shadow 

caused by a human hand covers the PV panel whereas in the 

second case, a piece of sacking covers the PV panel.  

As Figure 7 (a) illustrates, the brightness and color distortion 

values of the shaded region are calculated using the proposed 

image processing method based on Matlab. The results are as 

follows: 

Bshaded = 0.1068, CDshaded = 0.2388 

By applying rule number 6 on these values of brightness and 

color distortion, the output will be a shadow.  

For the second input image in Figure 8 (b), the brightness and 

color distortion values are: 

Bshaded =0.1323, CDshaded = 2.7019 

By applying rule number 4 on these values of brightness and 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 6. The input and output membership functions of the pro-

posed fuzzy logic method. 

 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

5  

color distortion, the output will be an object.  

     The fuzzy logic decision-making approach defines the 

type of the shaded region on the PV input image, and then this 

information will be used as an input parameter for active by-

pass circuit. In addition, when the type of shadow has been 

determined the cleaning purpose will be used. If the shadow 

is an object, it should be removed. On the other hand, if a 

shadow is detected, it should be cleaned using any cleaning 

method.   

 

 
 

IV. PROPOSED CONFIGURATION OF THE 
ACTIVE BYPASS CIRCUIT AND ITS 
CONTROL PRINCIPLE 

Fig.8 illustrates the proposed configuration of the active 
bypass circuit where a chopper circuit consists of capac-
itor, inductor, and MOSFET are connected in parallel 
with each PV module. The arrangement forms a buck-
boost converter which converts the fixed DC input volt-
age to the variable DC output voltage directly. Fig.9 
shows the gate signals for the multistage chopper cir-
cuit. 
 
 
 
 
 
 
 
 
 

TABLE 1 

Fuzzy Logic Rules. 

Rule No Rules 

1 Bshaded = b & CDshaded = b , then output = object 

2 
Bshaded = b & CDshaded = m , then output = object 

3 Bshaded = b & CDshaded = s ,  then output = object 

4 Bshaded = m & CDshaded = b , then output = object 

5 
Bshaded = m & CDshaded = m , then output = shadow 

6 
Bshaded = m & CDshaded = s , then output = shadow 

7 
Bshaded = s  , then output = shadow 

 

 

 

Fig. 7. The proposed shadow detection method. (a)Shadow region 

 

 
 
 
 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
Fig. 7. The proposed shadow detection method. 

(b) Object region. 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

 

 

 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

6  

 
 
 
 
 
 
 
 
 
 
 
 
 

Fig. 8. The proposed configuration of the active bypass circuit. 
 
 
 
 
 
 
 
 
 
 
 
 
 

Fig. 9. The gate signals for the multistage chopper circuit. 

The generation control voltage, for each PV module 

(VPVk), depends upon the switching frequency of the 

buck-boost converter which can be given as follow: 

VPV1 : VPV2 : ………. : VPVk  = D1 : D2 : ………. : Dk        

(6)                              

The switching duty ratio is given by: 

         

 

     

      

                                                                                  (7) 

      

    (7) 

 

 

Where Dk is the switching duty ratio, Tk(off) is the OFF-

time of switch gk, Ttotal is total switching interval, K is the 

number of PV modules, VPVk is the generation control 

voltage and Vo is the output voltage.   

In order to benefit from the active bypass features, with 

accreditation of shadow detection method which based 

on digital image processing technique, this approach 

uses a pulse width modulator (PMW) to control the ac-

tive bypass circuit by defining the gate signals of the 

MOSFET. The proposed PWM for the active bypass cir-

cuit is shown in Fig.10 where the PWM can be con-

trolled by the input value of brightness and color dis-

tortion of the shaded area which is obtained from the 

shadow detection method based on digital image pro-

cessing techniques. Fig.11 shows the gate signals g1 and 

g2 for the active bypass circuit.  

 
 
 
 
 
 
 
 
 
 
 

Fig. 10. The proposed pulse width modulator. 

 
 
 
 
 
 
 
 
 
 
 

Fig.11. The gate signals for the proposed active bypass circuit. 

V. FUZZY LOGIC CONTROL TECHNIQUE FOR 
TRACKING MPP 

   In order to track the MPP of the proposed grid-

connected PV solar system, this paper uses a fuzzy 

logic controller (FLC). This type of controller achieves 

a quick location to the MPP by controlling the gate sig-

nals of the MOSFET of the proposed active bypass cir-

cuit. Fig.12 shows the proposed simulation model of 

the FLC based on Mamdani method using the 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

7  

Matlab/Simulink/Fuzzy Logic Toolbox. Fig. 13 illus-

trates the inductor current (IL1) waveform in the active 

bypass circuit when the signal has been received from 

the fuzzy logic controller.   

 

 

 

 

 

Fig. 12.  Block diagram of the proposed FLC. 

 
 
 
 
 
 
 
 
 
 

 

Figure 13: The inductor current waveform in the active bypass 

circuit. 

 

The inputs of the proposed FLC are the PV array output 

power and current: 

   

 

where 

∆Ppv is the change in the output power of the PV array, 

∆Ipv is the change in the output current of the PV array, 

and j is the sampling instant.  

The output of the proposed FLC is the duty ratio value 

(D) of the active bypass circuit which controls the ON-

OFF time of switching period for the gate signal of each 

MOSFET. The input membership functions, the change 

in power and change in current of the PV array, and the 

output membership function of the proposed fuzzy 

logic method are presented in Fig.14. The fuzzy logic 

rules are shown in Table 2, where 16 fuzzy control rules 

are used to get the desired value of the MPP. The sur-

face view for the proposed FLC is shown in Fig.15. 

where: 

NB: Negative Big, NS: Negative Small, PB: Positive Big, 

and PS: Positive Small.  

Fig. 14.  The input and output membership functions of the pro-

posed FLC. 

TABLE 2 

Fuzzy Logic Rules. 

 

 

 

 

 

 

 

 

 

 

 

Fig. 15. Surface viewer for the proposed FLC. 

The CSI output current is illustrated in Fig. 16. In 
order to reduce the output harmonics on the AC side, 
the LC filter is connected between the CSI and the grid. 

The proposed fuzzy logic based MPPT controller 
includes three basic components, fuzzification module, 
inference engine and defuzzification module as shown 
in Fig. 17. The fuzzification makes it possible to pass 
fromthe crisp variables to the linguistic variables. The actual 

voltage(Vpv) and current (Ipv) of the PV module can be meas-
ured continuously and the power can then be calculated  



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

8  

(Ppv = Vpv× Ipv). The control is determined on the basis of 

satisfaction of two criteria relating to the two input variables 

of the proposed controller, namely ∆Ppv and ∆Ipv at the sam-
pling instant j. 

 

 

 

 

 

 

 

 

 

 

 
Fig. 16. The CSI output current. 

 
 

The output is the duty ratio value (D) of the active by-
pass circuit which controls the ON-OFF time of the 
switching period for the gate signal of each MOSFET 

 

Fig. 17. Structure of fuzzy logic controller. 

 
The inference engine applies the rules to the fuzzy in-

puts (that were generated from the fuzzification process) to 

determine the fuzzy outputs. Therefore, before the rules can 

be evaluated, the crisp input values must be fuzzified to ob-

tain the corresponding linguistic values (that are necessary to 

determine the active or fired rules) and the degree to which 

each part of the antecedent has been satisfied for each rule.   

 

It is noticed that the inference methods provide a func-

tion for the resulting membership variable; it thus acts on 

fuzzy information. As the DC-DC converter requires a pre-

cise control signal D at its entry, it is the necessary to en-

visage a transformation of this fuzzy information into de-

terministic one, this phase is called defuzzification. Defuzz-

ification can be performed normally by two algorithms: Cen-

ter of Area (COA) and the Max Criterion Method (MCM). 

The most used defuzzification method is that of the determi-

nation of the center of area (COA) of final combined fuzzy 

set. The fin union of all rule output fuzzy set using the max-

imum aggregation method. 

To validate the performance of the proposed PV solar sys-

tem shown in Fig.1, the maximum power point tracking de-

sign is simulated using a Matlab/Simulink. The circuit spec-
ifications are given in Table 3. The output power and 
current of the proposed PV solar system under  uniform 
condition (with irradiance level 500W/m^2) are shown 
in Fig.18 

 
 
 
 
 
 
 
 
 
 
 
 
 

(a) 
 

 
 
 
 
 
 
 
 
 
 
 
 

(b) 
Fig. 18. (a) PV output power. (b) PV output current under the uni-

form conditions. 

When the second PV module in Fig. 1 is partially shaded 

(with irradiance level 250W/m^2), the output power and cur-

rent of the proposed PV solar system are shown in Fig. 19. 

 
 

 

 

 

 

 

 

(a) 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

9  

 

 

 

 

 

 

 

(b) 

 Fig. 19. (a) PV output power. (b) PV output current under partially shaded 

conditions. 

 

 

 

 

 

Fig. 20. Block diagram of the grid voltage and current.  

 

TABLE 3 

The PV solar system specifications. 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

 

 

 

 

The comparison between the results of output power and 

current for the PV modules under the normal conditions and 

partial shading conditions with using the proposed MPPT 

control method is shown in Table 4. 

TABLE 4 

PV output power and current 

 

The AC grid voltage and current are shown in Fig. 21, 

where the two signals are out of phase. In order to make both 

signals in phase, a phase-locked-loop is used as shown in Fig. 

22. The ac grid voltage and current with using phase-locked-

loop are shown in Fig. 23. 

(a) 

        (b) 

Fig. 21. (a) The grid voltage. (b) The grid current. 

Fig. 24 shows the simulation result of the proposed 

MPPT control method under different shadow conditions. 

The power generated from the two PV modules under the uni-

form conditions at T1 is equal to 225W. When the second 

module is partially shaded at T2, the PV output power de-

creases to 158.77W. When the shadow is removed at T3, the 

PV output power increases again to the same value. This re-

sult shows that the proposed MPPT control method is stable 

under the different shadow conditions. 

PV output power without a shadow 225w 

PV  output power with shadow 182.35w 

PV  output current without a shadow 15A 

PV  output current with shadow 13.5A 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

10  

Fig. 23. The AC grid voltage and current with using phase-locked-loop. 

Fig. 24. Simulation results of the proposed MPPT under different shadow 

conditions. 

 

VI. CONCLUSION 

With the increased demand for renewable energy resources 

especially the solar energy, it is important to preserve the per-

formance of the PV array of the solar system on its maximum 

power point under the partial shading conditions. In this the-

sis, the partial shading problem of the photovoltaic solar sys-

tem has been investigated and addressed. Specifically, the 

global maximum power point has been tracked. In fact, the 

conventional algorithms cannot solve the problem of multiple 

peaks that appear in the P-V curve when the solar irradiance 

on the PV array is changed as a consequence of partial shad-

ing. To overcome that problem, this thesis has used an artifi-

cial intelligence algorithm that based on a fuzzy control algo-

rithm. Furthermore, this thesis presented an effective process 

to detect the shaded region on a PV module. This process uses 

a digital image processing technique to tack an image for the 

PV. module, and then this process determines the existence of 

a shadow or not on PV array. Finally, using a fuzzy design 

making method, this thesis has introduced a practical method 

to discriminate the type of this shadow. The results obtained 

from the shadow detection method have been used to design 

a pulse width modulator which can enable the gate drive of 

the active bypass circuit. This active bypass circuit ensures 

that every PV module on the solar system is working at its 

maximum operating point. This should increase the power 

generation of the PV solar system. Consequently, an effective 

method to track the maximum power point under the partial 

shading conditions has been developed based on the combi-

nation between the fuzzy logic technique and active bypass 

circuit which ensured a unique MPP for the solar system. In 

order to connect the PV array of the proposed solar system to 

ac grid, a current source inverter (CSI) has been used with a 

double tuned resonant filter which showed up eliminations in 

the undesirable harmonics on the DC side. In addition, an LC 

filter has been used in AC side to reduce the switching har-

monics 

The simulation results showed an improvement in 

the performance of the proposed MPPT method under partial 

shading conditions. Hence, the combination of active bypass 

circuit and the image processing technique, which relies on 

fuzzy logic decision-making approach, provides an effective 

approach to detect the shaded region on the PV module. In 

addition, this combination ensures a unique MPP of the pro-

posed PV solar system. The proposed fuzzy control system 

can reach the maximum power in a short time and can main-

tain its stability under any partial shading conditions. 

REFERENCES 

[1] S. Silvestre, A. Boronat, and A. Chouder, “Study of 
bypass diodes configuration on pv modules,” Applied 

Energy, vol. 86, no. 9, pp. 1632 – 1640, 2009. 

[2]  H. Ziar; S. Mansourpour; E. Afjei; M. Kazemi, “By-pass 
diode characteristic effect on the behavior of solar PV 

array at shadow condition”,3rd Power Electronics and 

Drive Systems Technology (PEDSTC), pp. 229 – 233, 

2012. 

[3]  H. Ziar, M. Nouri, B. Asaei, and S. Farhangi, “Analysis 
of overcurrent occurrence in       photovoltaic modules 

with overlapped bypass diodes at partial shading,” IEEE 

J.Photovolt., vol. 4, no. 2, pp. 713–721, Mar.  

2014. 

[4]  M. A Hasan; S. K Parida, “Effect of nonuniform 
irradiance on electrical characteristics of an assembly of 

PV panels”, IEEE 6th India International Conference on 

Power Electronics (IICPE), pp.1-3, 2014. 

[5] R.Dash, S.C.Swain, P.C.Panda, “A Study on the 
Increasing in the Performance of A Solar Photovoltaic 

Cell During Shading Condition”, IEEE International 

Conference on Circuit, Power and Computing 

Technologies (ICCPCT), 2016. 

 

[6] T. Shimizu, M. Hirakata, T. Kamezawa, and H. 
Watanabe, “Generation control circuit for photovoltaic 

modules,” Power Electronics, IEEE Transactions on, vol. 

16, no. 3, pp. 293–300, May 2001. 

[7] T. Shimizu, O. Hashimoto, and G. Kimura, “A novel 
high-performance utility-interactive                    

photovoltaic inverter system,” Power Electronics, IEEE 

Transactions on,   vol. 18, no. 2, pp. 704-      

      711, Mar 2003. 



M. Ibbini and A. Al-Obeidallah/ Maximum Power Point Tracking Control Method for Grid -Connected Photovoltaic System under Partially 

Shaded Conditions (2020)  

 

  

11  

[8] M. Karakose, M. Baygin, "Image Processing Based 

Analysis of Moving Shadow Effects for Reconfiguration 

in PV Arrays, "Energycon, Dubrovnik, Croatia, May 

2014. 

[9] M. Karakose, K.Firildak, "A Shadow Detection Approach 

Based on Fuzzy Logic Using  Images        Obtained from 

PV Array ", 2015 6th International Conference on 

Modeling, Simulation, and Applied Optimization 

(ICMSAO), pp. 1-5, 2015. 

[10]   R. Giral, C.A. Ramos-Paja, D. Gonzalez, J. Clavinet,       

A. Cid- Pastor, and L. Martinez Salamero,          

“Minimizing the effects of shadowing in a PV module            

by means of active voltage sharing,” IEEE Int. Conf. on 

Ind. Technol. (ICIT), pp.943-948, March 2010. 

[11] R. Giral; C. E. Carrejo, M. Vermeersh, A. J. Saavedra-   

         Montes, C. A. Ramos Paja, 

“PV field distributed maximum power point tracking   

         by means of an active bypass  converter,” Int. Conf.   

         Clean Electrical Power (ICCEP), pp. 94 – 98, 2011. 

[12]  Y. Benezeth, P.-M. Jodoin, B. Emile, H. Laurent, and C. 

Rosenberger, “Review and evaluation of commonly-

implemented background subtraction algorithms,” in 

2008 19th International Conference on Pattern 

Recognition, 2008, pp. 1–4.  

[13] L. Xuan, Z. Hong, “An improved canny edge         

detection algorithm,” In 8th IEEE International         

Conference on Software Engineering and Service         

Science (ICSESS), pp. 275 - 278, 2017. 

[14] I. Ahmad, I. Moon, S. Joo Shin, “Color-        to-grayscale 

algorithms effect on edge detection - A comparative 

study,” International Conference on Electronics, 

Information, and Communication (ICEIC), pp.1-4, 2018. 

[15] S. K. Pal and R. A. King, “On edge detection of X-ray 

images using fuzzy sets,” IEEE Trans. Pattern Anal. 

Mach. Intell., no. 1, pp. 69–77, 1983.  

[16] V. Torre and T. A. Poggio, "On Edge Detection," in 

IEEE Transactions on Pattern Analysis and Machine 

Intelligence, vol. PAMI-8, no. 2, pp. 147-163, March 

1986. 

 [17] J. Canny, “A computational approach to edge 

detection,” IEEE Trans. Pattern Anal. Mach. Intell., no. 

6, pp. 679–698, 1986.  

[18] H. Wechsler and M. Kidode, “A new edge detection 

technique and its implementation,” IEEE Trans. Syst. 

Man. Cybern., vol. 7, no. 12, pp. 827–836, 1977.  

[19] A. C. Bovik, T. S. Huang, and D. C. Munson, “The effect 

of median filtering on edge estimation and detection,” 

IEEE Trans. Pattern Anal. Mach. Intell., no. 2, pp. 181–

194, 1987. 

[20] A. L. Kabade, V.G. Sangam, “Canny edge detection 

algorithm,” IJARECE, vol. 5, no 5, pp. 1292 – 1295, May 

2016. 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.R.%20Giral.QT.&newsearch=true
http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.C.%20E.%20Carrejo.QT.&newsearch=true
http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.M.%20Vermeersh.QT.&newsearch=true
http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.A.%20J.%20Saavedra-Montes.QT.&newsearch=true
http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.A.%20J.%20Saavedra-Montes.QT.&newsearch=true
http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=%22Authors%22:.QT.C.%20A.%20Ramos-Paja.QT.&newsearch=true
http://ieeexplore.ieee.org/document/6036360/
http://ieeexplore.ieee.org/document/6036360/
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6029683
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6029683