Power Losses Reduction of Solar PV Systems Under Partial Shading Conditions Using Re-Allocation of PV Module-Fixed Electrical Connections

The paper gives a review of the investigation, which is being imported under different shading scenarios on the solar photovoltaic (PV) array arrangements. Under non-uniform irradiations, the authors pursue voluminous interpretation to scrutinize the mismatch power losses (PL) in PV array systems. In addition to power, the partial shading also reveals non-linearity along with multiple maximum power points (MPP) on performance curves i.e. power-voltage (P-V) and current-voltage (I-V). The inspection of the optimal layout of PV modules during the study helps us to extract maximum power and reduce the number of power peaks, when arranged in an array under partial shading conditions (PSCs). In the vicinity of PV array configurations such as series-parallel (SP), honey-comb (HC), bridge-link (BL), total cross-tied (TCT), and hybrid series-parallel-Total cross-tied (SP-TCT), bridge link-honey comb (BL-HC), and bridge link-total cross-tied (BL-TCT) are considered to investigate the performance under shadowing conditions. Moreover, Latin square (LS) puzzle is introduced to reconfigure the PV array and extensive comprehensive comparison with conventional is presented and entitled as “LS-TCT” configuration. The MATLAB/Simulink environment helps in modelling all the considered PV array configurations. The recommended LS-TCT configuration is turned out to be superior (for MATLAB/Simulink study) among all configurations during PSCs in terms of location of global maximum power point (GMPP), minimized PL and improved fill factor (FF). To show the superiority of proposed Shape-do-Ku (SPDK) puzzle based configuration, an experimental comparison is shown with conventional TCT and LS-TCT, Su-do-Ku (SDK) puzzle based configurations under new shading pattern-4. In addition, presented experimental study is validated the results obtained during MATLAB/Simulink study.


I. INTRODUCTION
In this era, when the repository of fossil fuels is a deficit and limitedly constrained, the rejuvenated analyst needs to explore productive renewable energy (RE) sources. In the The associate editor coordinating the review of this manuscript and approving it for publication was Salvatore Favuzza . framework of the optimum availability of bio-fuel cell energy, wind turbines, and PV systems are emerging RE sources of energy without denying the fact that they have their own limitations towards environmental aspects [1].
PV technology confronts enormous issues due to numerous known and un-acquainted causes' e.g. mal-function and climatic contaminants. In the modern consequence, the restrictions associated with environmental are rapidly growing due to dust, which is accumulated on the panel surface. The reasons behind the formation of diverse shadows on the surface of PV plants may be due to clouds shifting, neighboring trees, and upraised buildings especially in metropolitan areas [2]. Recently, behavioral enhancement in PV technology has been a quite fascinating area of research with the amendment in electrical connections of PV modules to enhance PV system performance [3]. The PV modules of an array are organized in series, parallel, SP, BL, TCT, and HC configurations [4], but the rearrangement techniques of PV modules are adopted for the performance enrichment in contrast with the conventional ones. The root cause and impact of PSCs on PV system performance are shown in Fig. 1 as, Presently, researchers are exploring ideas to mend the performance of PV systems under PSCs by using mathematical puzzle-based arrangement of PV modules (m × n) in an array. Various applicable methods are available in the current circumstances to reconfigure PV array systems by altering physical position of module with fixed electrical connections.
The module arrangement in a PV array exists in series and/or parallel to fulfill the load power requirement. In times, when one or more solar panels are shaded, substantially PV performance decreases. In [5], the authors have done comprehensive investigation based on performance outcomes e.g. PL, FF, and GMPP locations under the four kinds of shading instances such as long narrow (LN), long wide (LW), short narrow (SN) and short wide (SW). Rearranging the traditional TCT arrangements based on Su-Do-Ku game puzzle for extensive analysis. The maximum power produced by Su-Do-Ku configured structure is 4532W with a power enhancement of 26.1% over TCT configuration. In experimental approach, local and GMPPs are recognized and validated through estimated power as 150W under shading circumstances with the MATLAB/Simulink study [6]. Three series-connected PV panels are believed to determine the effect of irradiation under a non-uniform state at the GMPP site and power output is observed as 165W [7]. In order to achieve the MPPT during the shaded SP, TCT, and BL arranged PV modules (2 × 4), an experimental and MATLAB/Simulink studies are conducted for validation outcome. It is determined that the TCT arrangement has the best results compared to other modules with a power output of 678.40W [8] and when arranged in 4 × 4 array produces a power of 2.86KW [9]. The authors of [10], [11] have developed 3 × 3 SP configured PV array to demonstrate the shadow's effect on the P-V curve. Moreover, GMPP location is observed as 40W.
In [12], three shading circumstances i.e. 100%, 66.66%, and 33.3% insolation levels, the author inspected the series and parallel arrangements with grid integration. The outcomes for enhanced FF of 64.54%, low mismatch losses (MML), and GMPP location is noted as 290.5W. For testing of 230W PV panels (eight numbers) arranged in series connections and conducted under three shading test cases in both simulation and experimental aspects. The placement of the bypass diode with the PV system is evaluated during the research to reduce the shadow effect [13]. The optimal interconnection technique is used to scatter the shadow impact on the TCT configured PV array and compared with the outcomes with Su-Do-Ku puzzle in terms of enhanced power of 4802W [14]. For comprehensive research under PSCs, PV array settings defined as series, parallel, SP, TCT, BL, HC, and suggested new PV array setup are regarded in [15]. Compared to others, the proposed 'novel' configuration has a stronger performance. An improved Su-Do-Ku model is obtained from existing Su-Do-Ku PV model in [16] to analyze the efficiency under four shadowing instances such as SW, LW, SN and LN. In addition, Su-Do-Ku based configuration has the highest power as 1250W and lowest PL among all the experiment instances [17].
In [18], the authors have analyzed TCT and Su-Do-Ku puzzle based reconfigured TCT (RTCT) array systems under passing clouds as a shadow impact on the 6 × 6 size PV system with estimated GMPP as 1160W. Under the distinctive shadow motifs, the performance of standard TCT and Su-Do-Ku puzzle-based RTCT setup is compared, and found enhanced GMPP as 2278W [19]. Results from TCT, SP-TCT, and Su-Do-Ku connections are analyzed. In which, Su-Do-Ku configured PV array have stronger findings as improved FF and less numbers of MPPs on P-V curve. Under shading cases i.e. SW, LW, SN, and LN, the production strength of a 7×5 PV system increased in terms of maximum power to 73.55W by using various settings such as electrical panel setup and physical re-allocation of PV module and fixed electrical connection [20]. Moreover, Futoshiki arrangement of 5 × 5 PV system is investigated and it has been found 64.87W as power output [21]. Optimum PV panel arrangement in an array are chosen and contrasted under the predefined shading impacts with SP and TCT interconnections. The TCT has highest power as 4419W [22]. In this study [23], author proposed a 6 × 4 MATLAB/Simulink model of an array to implement different configurations i. e. TCT, and hybrid SP-TCT, BL-TCT, BL-HC and novel structure (NS) at four distinct insolation levels such as 350W/m 2 , 500W/m 2 , 800W/m 2 , and 1000W/m 2 . The experimental analysis leads the author to observe that number of GMPPs compared to TCT. In [24], author have done both simulation and experimental analysis of 6 × 6 size PV array arranged in TCT and magic square (MS) arrangements. The authors have observed that the highest GMPP location as 300W at irradiation levels from 500-1000W/m 2 in the case of MS configuration.
In [25], the author has considered distinct and rational accelerating shading methodologies such as left to right, bottom to top and diagonally multi-story building pattern for amending the behavioural evaluation of TCT configuration over the LS puzzle pattern. The performance parameters such as FF, PL, and GMPP are visualized for TCT and LS-TCT configurations and thus found LS-TCT to be superior with 78.7%, 330W, and 2279W respectively. The author of [26]  In [28], the authors have given a unique approach of Skyscraper puzzle-based 9 × 9 PV configuration and tested under LN, SW, SN and LN shading scenarios. It is found that the GMPP location is higher side as compared to TCT, Dominance square and Su-Do-Ku (22.36%, 43.36%, and 39.31%) arrangement respectively. The authors in [29], investigated Su-Do-Ku based BL-TCT topology and observed that the power produced as 44.31W by the proposed configuration at different irradiations. The authors in [30] has investigated PV system performance under irregular solar irradiance (400-1000W/m 2 ) on conventional PV configurations i.e. SP, HC, BL, and TCT, reconfigured method (RM). The author [31] have discussed wide shade dispersion and introduced FRA, SMO algorithm, and RAO algorithm using the same which has allowed him to magnify power by 13%, 11%, and 9% respectively when contrasted with 9 × 9 size TCT, CS, and GA topologies. Under non-uniform irradiation levels of 200W/m 2 , 400W/m 2 , 500W/m 2 , 600W/m 2 and 900W/m 2 different performance parameters i. e. FF, ML, %PL, and %PE are observed and thus concluded full reconfigured array with higher power and reduced MPP. The game puzzles are used to reconfigure the conventional TCT PV array configurations. In this paper, the PV array configurations such as LS-TCT is considered. The property of shade dispersion is always vary based on set of integer numbers to design the game puzzle. 9 × 9 size PV array is investigated to show the comparison with existed TCT configuration under four shading scenarios as SW, LW, SN and LN. The minimized PL are observed as 6.75W, 3.98W, 4.68W and 6.95W respectively [32]. Another technique has proposed [33] hybrid reconfiguration algorithm, this technique investigate which step matrix is more efficient. The presented technique decrease the effect of partial shading in the power output and circuit. In [34], the Coyote Optimization Algorithm (COA) is used to reconfigure the PV array and compared to TCT, Su-Do-Ku, flower pollination algorithm (FPA), butterfly optimization algorithm (BOA), and marine predictor algorithm (MPA) PV array configurations. Under four shading scenarios (SW, SN, LW and LN), the COA-based GMPP configuration is found to be 11kW, 1.08kW, 1.45kW and 1.560kW respectively. Another Luo Shu technique [35] has been used for prevent MML, output power and other topology of PV array under shading area. Comprehensive comparison is carried out between TCT, Su-Do-Ku, Dominance square (DS), Competence square (CS) and proposed Luo Shu PV array configurations under shading scenarios. Through, that obtained I-V and P-V curves clearly shown higher performance of Luo Shu PV array.

A. NOVELTY OF WORK
The goal of this paper is to organize the position of modules physically without altering the electrical links to improve PV performance under PSCs. The game puzzles e.g. LS-TCT, SDK, SPDK configurations are tested to shown an efficient result due to subsequent reasons.

II. PV SYSTEM TECHNOLOGY AND ARRAY TOPOLOGIES
Strategies assumed in the proposed work for modelling of PV system is clarified in the section given below,

A. MODELING OF THE PV SYSTEM
The arrangement of PV modules connected into SP produces high power generation as presented in Fig. 2. Solar PV voltage of a unit cell (V C ) depends upon solar irradiation and is the function of current (I SC ) which can be conveyed through Eq. (1) as, The solar irradiation level (S C ) and ambient temperature (T a ) are the two factors varying the operating temperature (T C ) of PV cell. Voltage and current correction factors i.e. C TV and C TI respectively shown in Eq. (2) as, C SV & C SI are two correction factors which are responsible for the outcome of irradiation level (S x ). The correction factors C SV , for voltage and C SI , for photocurrent can be conveyed through in Eq. (3) as, where, S c and S x referred to the solar and actual irradiation level respectively. The actual values referring to PV cell such as photocurrent i. e. I phx and its voltage i. e. V cx can be determined through correction factors C TV , C TI , C SV and C SI can be expressed in Eq. (4) as Commercially available PV module specifications at standard test scenarios are considered for MATLAB/Simulink modelling to investigate PV array systems under PSCs as,  Fig. 3(a)-(c) represents conventional SP, BL, and HC configuration for an array of size 4 × 4 of PV panels where its conduct is measured in relation to GMPP and FF. Fig. 3(d) represents TCT arrangement, for array size 4×4 of PV panels, which is obtained by ties connected across individual rows of interconnections to produce high power. The recombination of classical SP, BL, and HC along with TCT topology, represents the hybrid topology of PV array and is obtained in Fig. 3(e)-(g) below. LS is an ancient puzzle where users try to find out how many integer numbers can be arranged in a specified number of rows and columns. Moreover, each symbol appears only once in each row and column in the matrix and developed by Leonard Euler. Property of such a puzzle with higher shading dispersion during the reconfiguration of PV array under the PSCs [32].

2) LS PUZZLE BASED PV ARRAY CONFIGURATION
The property of shade dispersion is always vary based on set of integer numbers to design the game puzzle. So, the placement of LS puzzle having different properties with various placement possibilities of integer number (1)(2)(3)(4). The methodology to reconfigure PV array based on LS puzzle is shown in Fig. 3(k).

3) SU-DO-KU PUZZLE BASED PV ARRAY CONFIGURATIONS
For shade dispersion, this SDK puzzle is introduced first to reconfigure PV array in 2013 [5] and tested under PSCs. The 4 × 4 size PV array is reconfigured using set of rules with integer numbers (1)(2)(3)(4) in each row-column of the matrix. The potential electrical connections and design methodology are shown in Fig. 3(i) and Fig. 3(l), respectively.

4) SHAPE-DO-KU PUZZLE BASED PV ARRAY CONFIGURATION
The SPDK puzzle is existed in various sizes (4×4, 5×5, 6×6) with symmetrical properties and it has no technical relation to the SDK puzzle. But, it is a version of the LS puzzle with the only restriction that each number appear at least single time in each column and row [36]. The PV array is reconfigured in this paper using a 4 × 4-size SPDK puzzle. Fig. 3(j) and Fig. 3(m) demonstrate the electrical connections and design methodology, respectively.

III. ANALYSIS OF PERFORMANCE PARAMETERS AND SHADING PATTERNS
To prove any method as a superior technique, it's necessary to evaluate its performance parameters. With this notice, in this article authors have considered different performance parameters to assess the superiority of proposed PV array reconfiguration technique. The detailed analysis of these parameters are as follows:

B. ANALYSIS OF SHADING SCENARIOS
The paper demonstrates eight forms of arrangements i.e. SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT and LS-TCT in each case with three cases in each shading pattern. In the article, three distinct types of shading patterns are distinguished with the help of shadow movements. The irradiation levels such as 1000W/m 2 and 500W/m 2 are considered during the study.

1) SHADING PATTERN-1
It constitutes three cases-1, 2, and 3 with different shadow movements as presented in Fig. 5 as, Three PV modules of the first, second, and third-row receive uniform solar irradiation of 1000W/m 2 . Moreover, in the fourth row, two modules receive irradiation of 500 W/m 2 and the leftover two modules receive 1000W/m 2 of solar irradiation. So, to evaluate the current produced in the first, second, and third row of 4 × 4 PV array is denoted in Eq. (8)- (13). The current generated under shading case-1 as, Similarly, the current generated under case-2, The generated current in fourth row is calculated as, Moreover, the current generated under shading case-3 as, 2) SHADING PATTERN-2 The shadow is presumed to be approaching towards right from the leftmost bottom two modules progressing upwards as depicted by cases 1-3 of Fig. 6 as, Theoretical analysis of generated current under shading cases 1-3 can be carried out and expressed in Eq. (14) 3) SHADING PATTERN-3 Considering shading pattern-3, it is observed that there is an upward movement of shadow as we are switching from pattern 1 to 3 of PV array and in each arrangement. The left to right shading on PV modules is shown by cases 1-3 of Fig. 7

IV. RESULTS AND DISCUSSION
With the due effect of various PSCs, the following investigations have been demonstrated on various PV array configurations to estimate the extensive performance.

A. P-V AND I-V CURVES AT STC
The single PV module's P-V and I-V curves at STC and insolation levels with power at GMPP as 2610W through MATLAB simulations are presented in Fig. 8 as, VOLUME 9, 2021

B. EFFECT ON PV ARRAY CONFIGURATIONS UNDER SHADING PATTERN-1 1) P-V AND I-V CURVES
The I-V and P-V curves of all the possible topologies of PV array undergoing three distinct types of shading patterns from bottom to top including three cases in each pattern (left to right) from cases 1-3 is demonstrated and differentiated through Fig. 9. From Fig. 9(a), it is clearly inferred that two MPPs i. e. GMPP and LMPP are existing on the P-V curve for all the topologies. Both MPP's are found to be far apart from each other and lead to a small increase in the consequences of partial shading however the global point for LS-TCT is smooth for case '5a'. The power obtained at GMPP is noticed as 2053W, 2134W, 2091W, 2169W, 2169W, 2118W, 2118W, and 2368W for the SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT, and LS-TCT topologies as presented in Fig.10. The P-V curve for the case referred to in Fig. 5(b)-(c) are presented in Fig. 9(b)-(c) respectively. With reference to the shading instance of Fig. 5(c), it has been inspected through Fig. 9(c) that in classical and hybrid topology, the local and global MPP are far apart from each other. The obtained power at GMPP for these two cases are identified as 1988W, 1960W, 1942W, 1942W, 1942W, 1942W, 1958W, 2333W for case '5b ' and 1942W, 1942W, 1942W, 1942W, 1942W, 1942W, 1942W, 2247W for case '5c' respectively as shown in Fig. 10. The current, voltage, and power of LS-TCT interconnections are shown in Table 2. The produced power is incremented in few cases of distinct shading scenarios for LS-TCT configurations. Due to the rearrangement of modules, the shadowing consequence of shading pattern-1 gets dispersed over the PV array with an increment in power.

2) POWER LOSS
In each case, the LS-TCT puzzle pattern displays better performance in comparison to SP, BL, HC, TCT, BL-HC, BL-TCT, and SP-TCT when subjected to shade pattern-1. The power losses are expressed using a bar graph and  through tabular data presented in Fig. 11 and Table 2 respectively. The PL are observed to be very less in the LS-TCT puzzle configuration under all the shading conditions for pattern-1.

3) FILL FACTOR
The effect of shade gets escalated with its movement from one end to another on the PV Array resulting in a decrease in FF. Through shading pattern-1 in Fig. 5(a)-(c), it is constituted as a bar graph and presented through Fig. 12. Values corresponding to FF are represented in Table 2 for all the configurations considered in shading pattern-1. It is clearly inferred that FF for LS-TCT topology is superior for the shading pattern-1.

C. EFFECT ON PV ARRAY CONFIGURATIONS UNDER SHADING PATTERN-2 1) P-V AND I-V CURVES
In cases '1-3' of shading pattern-2, the P-V trait are visible in Fig. 13 of the PV array. On the P-V graph, various MPP's are observed concerning the case represented through   Fig. 6(a) -(c). It has been inferred from the graph that local and global MPP are far apart from each other. It is observed with reference to the shading case in Fig. 6(a) that the effect of shading gets escalated as we move on from cases '1-3' on the P-V curve. The obtained power at GMPP is noticed as 1994W, 2010W, 2037W, 2059W, 2059W, 2059W, 2059W and 2108W for SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT, and LS-TCT interconnection respectively. For the shading cases presented by Fig. 6(b) and Fig. 6(c), the P-V plot is displayed in Fig. 13(b). The power obtained at GMPP for instance '2' and '3' is observed as 1686 W, 1721W, 1703W, 1725W, 1725W, 1725W, 1725W, 2004W and 1379W, 1379W, 1379W, 1379W, 1379W, 1379W, 1379W, 1892W for SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT and LS-TCT framework respectively as shown in Fig. 14 in the form of bar graph. The power and voltages at GMPP are the parameters for comparison of distinct forms of PV array framework and results are presented through Table 3. The procured result shows that the LS-TCT puzzle pattern generates leading power in comparison to eight 4 × 4 other topologies. The current, voltage, and power of LS-TCT topology are shown in Table 3. The array's production is incremented in few cases of distinct shading scenarios after the reordering of modules in accordance with LS puzzle pattern.

2) POWER LOSS
In each case of shading pattern-2, the LS-TCT puzzle pattern unveils better performance in comparison to all eight 4 × 4 interconnected topologies. The power losses are expressed using bar graph by Fig. 15, and through tabular form presented in Table 3 respectively. Power losses are found to be lower in the LS-TCT puzzle pattern under all the shading conditions.

3) FILL FACTOR
The effect of shade gets escalated with its movement from one end to another on the PV Array resulting in a decrease in FF. Through the cases of pattern-2 shown in Fig. 6 (a)-(c), it is constituted as a bar graph and presented through Fig. 16.
The values corresponding to it are represented in Table 3 for all the configurations considered in shading pattern-1. It is clearly inferred that FF for LS-TCT topology is superior for the shading pattern-2.

D. EFFECT ON PV ARRAY CONFIGURATIONS UNDER SHADING PATTERN-3 1) P-V AND I-V CURVES
In cases '1-3' of shading pattern-3, the P-V curves are presented in Fig. 17 of the PV array. On the P-V plot, various MPP's are observed with reference to the shading instance represented through Fig. 7(a) -(c). It has been inferred from the graph that local and global MPP are far apart from each other. It is observed in the shading instance referred to Fig. 7(a) that the effect of shading gets escalated as we move on from cases '1-3'on the power curve. The power demonstrated at GMPP is 1945 W, 1956 W, 1943 W, 1969 W, 1942 W, 1942 W, 1942 W, and 2004 W for SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT, and LS-TCT framework respectively. For the shading instance of Fig. 7(b), the P-V curve is presented in Fig. 17(b). It has been inferred through the graph that local and global MPP are far apart from one another. The power observed at GMPP is 1615 W, 1613 W, 1626 W, 1628 W, 1626 W, 1626 W, 1613 W, and 1715 W for SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT, and LS-TCT arrangement. For the shading instance of Fig. 7(c), the P-V plot is revealed in Fig. 17(c). The power observed at GMPP for this case is 1287 W, 1287 W, 1287 W, 1287 W, 1287 W, 1287 W, 1287 W, and 1368 W for SP, BL, HC, TCT, BL-HC, BL-TCT, SP-TCT and LS-TCT interconnection respectively as shown in Fig. 18 with bar graph representation.
The electrical performance parameters such as current, voltage, and power of LS-TCT arrangement are shown in Table 4. The PV array power generated is escalated in few cases considering distinct shading scenarios after the reordering of a unit (modules) in accordance with LS puzzle pattern.

2) POWER LOSS
In each case of shading pattern-3, the LS-TCT puzzle pattern unveils better efficacy among all interconnected topologies. The power losses are expressed using the bar graph presented by Fig. 19 and through Table 10 respectively. In addition, the power loss parameters is found to be least in the LS-TCT puzzle pattern under all shading conditions for pattern-3.

3) FILL FACTOR
The effect of shadow gets escalated with its movement from one end to another on the PV Array resulting in a decrease in FF. Through the three shading instances of shading pattern-3 in Fig. 7 (a)-(c), it is constituted as a bar graph and is presented through Fig. 20. The values corresponding to it are represented in Table 4 for all the configurations considered in shading pattern-3. It is clearly inferred that FF for LS-TCT topology is superior for the shading pattern-3.

E. RESULTS VALIDATION 1) ANALYSIS OF SHADING PATTERN-4
Under shading pattern-4, a comprehensive investigation to validate the performance of 4 × 4 sizes TCT, LS-TCT, SDK, and SPDK configurations is considered. This shading scenario is comprised of two realistic shading cases with irradiation levels such as 1000W/m 2 , 500W/m 2 and 300W/m 2 that are considered during the study, as shown in Fig. 21.

2) MATLAB/SIMULINK ANALYSIS: P-V AND I-V CHARACTERISTICS
A thorough comparative study of the achieved performance of 4×4 sizes of TCT, SDK, and SPDK configurations is being  considered. The MPP is located at 80W in ideal conditions. The behaviour of the obtained P-V curves for the regarded PV array connections is examined under two shading scenarios (pattern-4) and is depicted in Fig. 22(a)-(b). The TCT configuration suffered significant power losses due to a lack of cohesiveness between the PV module power maxima and array's GMPP. Under two non-uniform irradiance levels (1000W/m 2 and 500W/m 2 ), the GMPP of the TCT configuration is 47.08W in shading case-I. Furthermore, VOLUME 9, 2021 During a similar shading case-II, the highest power losses are observed to be 25.86W in TCT and LS-TCT configurations, with the lowest equal GMPP values being 54.14W. The GMMP locations for SDK and SPDK configurations are 61.81W and 61.95W, respectively. Overall, in the shading case-I study, SPDK has the best GMPP and the lowest power losses (18.05W), as well as an improved FF (0.714).
As shown in Fig. 23(a), the I-V characteristics of TCT, LS-TCT, SDK and SPDK configurations are tested under shading case-I (pattern-4). The highest maximum current (I m ) For shading case-II (pattern-4), the values of I m are 1.351A, 1.351A, 1.589A, and 1.350A for TCT, LS-TCT, SDK, and SPDK configurations, respectively. V OC values are observed to be 45.028V for all PV array configurations. It demonstrates that non-uniformity sun irradiance is more harmful to short circuit current than V OC .

3) EXPERIMENTAL ANALYSIS: P-V AND I-V CHARACTERISTICS
An experimental study on 4 × 4 size of TCT, LS-TCT, SDK and SPDK configurations is performed. Under ideal/uniform irradiance such as 1000W/m 2 , MPP is observed as 80W. The present study on PV system configurations is deliberated under similar shading cases I-II (pattern-4). Fig. 24 shows  the experimental setup and the major components of are as follows (i) sixteen PV modules arranged in 4 × 4 matrix (ii) resistive load (variable) and (iii) self-developed data logger to store real time voltage, current data to characterize PV system. This data logger system was developed with voltage and current sensors to measure real-time electrical data during experimentation. The open source Arduino System (ATmega-328 micro-controller) operated the system performance. The real time electrical data is stored in micro SD card for further investigation. Table-5 shows the specifications of commercially available solar PV modules used for MATLAB/Simulink modeling and experimental studies.
An extensive experimental study is performed and obtained electrical performance of TCT, LS-TCT, SDK and SPDK configurations is deliberated. The P-V curves for all four PV array configurations are obtained under shading cases I-II, shown in Fig. 25(a)-(b).
The TCT configuration is experiencing a large amount of shading losses due to lack of coherence between the power maxima point of PV modules in an array. In shading case-I, the GMPP of the TCT configuration is 45.77W at non-uniform irradiation levels such as 1000W/m 2 and 500W/m 2 . In addition, the GMPPs are found as 59.95W, 59.95W and 62.25W for LS-TCT, SDK and SPDK configurations respectively. The performance of SPDK is having best performance compared to all among PV configurations under PSCs.
Furthermore, it is observed that the TCT configuration has low value of GMPP as 52.62W under non-uniform irradiation levels such as 1000W/m 2 , 500W/m 2 , and 300W/m 2 . Under similar shading scenarios, LS-TCT, SDK Under shading case-II, the irradiation levels are considered as 1000W/m 2 , 500W/m 2 and 300W/m 2 for performance investigations. The short circuit current values are observed as 2.22A, 2.22A, 1.955A and 1.968A for TCT, LS-TCT, SDK, and SPDK configurations due to shading effect. In addition this, the value of FF is investigated as 0.526, 0.551, 0.6873 and 0.6982 respectively, SPDK configuration has best performance under distinguish shading scenarios.     study. In MATLAB/Simulink study, higher GMPP locations are observed as 63.89W and 61.95W for SPDK configuration compared to TCT, LS-TCT and SDK configurations during shading cases I-II (pattern-4) respectively.
In addition, highest power at GMPP is assessed as 62.25W and 61.95W for SPDK configurations compared to TCT, LS-TCT and SDK configurations during shading cases I-II (pattern-4) respectively during experimentation work.

5) POWER LOSS
Calculation of PL are carried out under shading case I-II (pattern-4). The obtained results are depicted in Fig. 28(a)-(b) for MATLAB/Simulink and experimentation work. The highest and lowest shading case-I is experienced by TCT and SPDK configurations as 32.92W and 16.11W respectively compared to LS-TCT and SDK configurations. Moreover, minimized power PL is found as 18.05W for SPDK configuration compared to TCT, LS-TCT and SDK.  Under experimentation work, the minimized PL are observed during shading case I-II are experienced for SPDK configuration as 16.11W and 18.05W respectively compared to LS-TCT and SDK configurations.

6) FILL FACTOR
FF assessment with dissimilarities observation found due to shading cases I-II (pattern-4) for the TCT, LS-TCT, SDK and SPDK configurations and statistically represented in figure 29. During MATLAB/Simulink study, SPDK has improved value of FF as 0.727 and 0.714 respectively. Moreover, the investigation of FF is carried out and observed highest as 0.806 and 0.698 during experimental study under similar shading cases I-II respectively.
Critical study on shading effect analysis on PV system is performed and all the quantitative outcomes are summarized in Table-5  The bypass diode (BPD) plays a huge role in power generation of a PV array. During the PSCs without bypass diode, the PV array carries a current of shaded modules generated current. As per the property of series connection, the series connected PV array produces less amount of current, due to this power generation capacity of PV array greatly reduces. On the other side, the bypass diode acts as additional path for the current flow in-case of a shaded condition. Thereby shaded PV modules current is by passed and remaining PV array in a position to carry actual current produced by un-shaded PV module. Due to this there exist two different levels of current in a Bypass diode connected system. Thereby it results multiple peaks in a PV array. However the system exhibits multiple peaks power generated capacity of a PV array will be high than without bypass diode connected system. To confirm the same, by considering the same PV modules, shading of TCT connected system, authors performed simulation under two different conditions and obtained results are table in Table to show clear understand on the impact of bypass diode over TCT connected PV system. From the table it can be observe that, there is less differ of power between those, however, if we implement the same to high rated system, different will be huge and it shows significant improve in power generation than without bypass diode connected system. In addition, the I-V and P-V characteristics have been plotted to shows its effect pictorially in Fig. 30(a)-(b). The quantitative values are depicted in Table-6.

V. CONCLUSION
This paper successfully investigated the performance of SP, BL, HC, BL-HC, TCT, SP-TCT, BL-TCT, and LS-TCT configurations using the MATLAB/Simulink models. The performance has been assessed by current, voltage, power at GMPP, FF, and PL. Due to dispersion in shading, LS-TCT configuration exhibited the lowest PL and increased FF values among eight electrical arrangements.
• The puzzle-based LS-TCT configurations are more productive leading to utmost PE of 8.4 %, power mismatch losses get reduced to 867W, and FF gets maximally escalated by 0.037 under shading pattern-1.
• The production of LS-TCT configuration has utmost PE by 27.11%, maximum redundancy in mismatch power by 1787 W, incremented FF by 0.32 compared to TCT (which has higher performance among all configurations, except LS-TCT) is observed for shading pattern-2.
• It has been acclaimed for the suggested puzzle-based configuration that the extracted output power is higher when study performed under PSCs.  He is currently working as an Associate Professor with the Department of Electrical Engineering, Chaitanya Bharathi Institute of Technology (CBIT), Hyderabad, India. He has completed his Postdoctoral Research Fellowship from the Institute of Power Engineering, Universiti Tenaga Nasional (UNITEN), Malaysia. Before to that, he worked as an Assistant Professor with the School of Electrical Engineering, VIT University. He has published more than 70 research articles in various renowned international journals. His research interests include the design and implementation of solar PV systems, renewable energy resources, power management for hybrid energy systems, storage systems, fuel cell technologies, electric vehicles, and smart grids. He has been acting as an Associate Editor of IET RPG, IEEE ACCESS, ITEES (Wiley), and Frontiers in Energy Research; the Section Editor of Energy and Sustainability (MDPI Publications); and a reviewer of various reputed journals.
HASSAN HAES ALHELOU (Senior Member, IEEE) is currently a Faculty Member with Tishreen University, Lattakia, Syria. He is with University College Dublin, Ireland. He has participated in more than 15 industrial projects. He has published more than 130 research articles in the high quality peer-reviewed journals and international conferences. His major research interests include power systems, power system dynamics, power system operation and control, dynamic state estimation, frequency control, smart grids, micro-grids, demand response, load shedding,