Programmable photovoltaic submodules for hotspot mitigation

ABSTRACT

This paper introduces a novel approach to conventional Photovoltaic (PV) modules that involves regulating the internal connections of the PV strings (or submodules) using a set of switches to isolate shaded submodules while leaving unshaded submodules unaffected. By extending the connection types of the PV submodules to include parallel or series via an external circuit, along with a set of switches, the output voltage and current can be adjusted without requiring additional PV modules to increase the output current. A microcontroller is employed to control the connections of the PV submodules to achieve three modes: series connection, parallel connection, or mixed connection. As a result, there is no need for a buck-boost converter, and the effects of partial shading are minimized, while the hotspot effect is mitigated without bypass diodes. A Matlab/Simulink model is utilized to validate the proposed configurations, and two customized hardware PV modules are fabricated to represent the conventional module and the proposed module. The results obtained from the hardware implementation support the concept of separating shaded PV submodules and regulating the internal connections of the submodules in different configurations. This feature of separating the unshaded submodules is implemented to enhance the efficiency of the PV modules.

KEYWORDS:

• Bypass diodes
• hotspots
• microcontroller
• partial shading
• Simulink

1. Introduction

1.1. Motivation

The impact of shading PV modules has been widely investigated, and many researchers have developed both static and adaptive solutions at the cell and module levels. An example of this solution is the use of BPDs, where the current is rerouted through the diode to prevent overheating of the shaded cell; however, this will reduce the PV module’s panel’s output power. The drawbacks of using these BPDs are the high cost, losses are not entirely prevented due to the voltage drop across the forward-biased BPDs, and BPDs may excessively heat up due to the high current passing across (Yaden et al. 2013). However, the losses due to the voltage drop across the BPDs can be mitigated by utilising bipolar junction transistors (BJTs) or metal oxide semiconductor field-effect transistors (MOSFETs). Another approach is the installation of one micro-inverter per submodule (Wirth and Wiesmeier C 2016) would be a valid solution, but it also holds some drawbacks, such as the reduced efficiency of the solar modules and the high expenses of implementing this concept.

This paper designs a smart architecture of a PV module to help mitigate hotspot formation, resulting in increased efficiency. Also, this work investigates the following areas: (1) adjusting the output current of the PV cells without the use of buck-boost converters, (2) minimising the impact of partial shading and mitigating the effect of hotspots without the use of BPDs, and (3) controlling the PV output voltage and current by only automatically adjusting the proper switches (Qin et al. 2015). In order to accomplish this, a simulation model Matlab/Simulink engineering calculations package, developed by (Chandani and Anamika 2014) using the ‘Solar Cell’ component to reflect the internal structure of the partially shaded PV module, was used. Verification tests are carried out for the photovoltaic module.

1.2. Literature review

Recently, the world has been gradually moving more towards dependence on renewable energy sources (Coşgun 2021). Solar energy is one of the renewable energy fields that has caught the attention of the world due to the straightforward installation of solar energy systems (Garcia-Barrientos et al. 2021). This is done through the use of Photovoltaic (PV) modules that convert solar irradiance into useful electricity (Bosman 2014). The dominant PV modules are fabricated of silicon crystal semiconductors composed of n-type and p-type layers that are stacked on each other (Lien et al. 2011). Once the sunlight strikes the module’s surface, DC electricity is generated as a function of the ambient conditions (Idzkowski, Karasowska, and Walendziuk 2020). One of the most common challenges is when PV cells receive non-uniform irradiance, which is interpreted in this article as partial shading. The consequences of shading generally are prevented by bypass diodes. A comprehensive review and highlights of recent advances in bypass diode applications are presented in (Vieira et al. 2020), and the shading effect and hotspot problem on PV modules are explained.

(Al-Chaderchi et al. 2018) conducted experimental (using the PV analyser) and simulation (MATLAB) work to study the impact of shading for different strings inside the PV modules under real outdoor environmental climate conditions. The outcome of his work was that increasing the number of bypass diodes in solar PV panels from 2 diodes to 4 diodes can improve the PV performance by 31% in the case of full cell shading. The work of (Wen et al. 2019) outlined the shading effect and hotspot problem on PV modules and developed an approach to mitigate the effects of partial shading by preventing the activation of bypass diodes and thereby reducing the chances of hotspots in mismatched cells. It also proposed a simple split submodules approach to investigate the effects of shading area and orientation on the electrical performance of the module. However, the main constraint was to further reduce shading loss, long submodules should be split into more submodules, and more bypass diodes should be included in the module.

Several designs have been suggested to decrease power loss caused by partial shading. However, these designs have not been evaluated for their ability to prevent hotspots and maintain the reliability of photovoltaic (PV) systems. (Swatilekha, Santosh, and Vinod 2023) examined the effectiveness of a hotspot prevention mechanism in both laboratory and real-world settings and indicated that this mechanism is not effective in preventing hotspots in the total cross-array configuration, which is considered the most effective configuration for minimising power loss due to mismatches.

There are lots of other studies on generating solar energy using PV modules, apart from the ones mentioned in the previous section. (Ge et al. 2022) investigated the current-voltage I-V characteristics of PV modules to determine whether there is a low resistance hotspot fault and carry out real-time control according to the judgement results. The work of (Tarabsheh, Akmal, and Ghazal 2021) discussed the effect of shading on the PV modules’ performance. However, the work of (Lee et al. 2022) demonstrated a quick recovery of the hotspot with on-site recovery technology compared to the factory recovery method, where the back sheet is removed and laminated to recover the PV module.

A non-invasive procedure was proposed by (Vélez-Sánchez et al. 2019) to estimate the parameters of the bypass diodes by solving a system of two nonlinear equations for PV modules using two experimental I-V curves. (Chao, Lai, and Chang 2019) developed a photovoltaic system that stores energy for use in direct current micro-grid systems as a smart maximum power point tracker system in shading conditions. Maximum power point tracking control is needed in the PV systems to harvest the maximum amount of energy from the sun; this control depends on changing the duty cycle of the DC-DC converter to obtain the maximum power point. Perturb and Observe algorithm is commonly used for maximum power point tracking; however, it fails during partial shading (Mahdi et al. 2020). By replacing the conventional Perturb and Observe algorithm with the proposed method, the losses during the partial shading are reduced significantly. As mentioned before, PV hot spots are considered one of the main reliability issues for PV modules. The common practice for mitigating this phenomenon is the adoption of the conventional bypass diode circuit, yet, this method does not guarantee a decrease in the temperature of hot-spotted PV cells. (Mahmoud and Ghadeer 2020) introduced a new current limiter circuit designed to reduce the current flow of PV modules that are experiencing mismatched conditions such as partial shading and hot-spotting. The proposed circuit is composed of an input buffer and an operational amplifier circuit, which regulates the current flow through integrated MOSFETs, allowing control over the amount of current flowing through mismatched PV sub-strings and subsequently increasing output power generation.

The electrical mismatch loss in PV cells by serially combining each unit module at various mismatch ratios was studied by (Park et al. 2022). Mismatch losses occur when the electrical parameters of one or more cells are different from the others, reducing the output power of the PV modules. The mismatch losses occur due to different causes, such as manufacturers’ tolerances, shadows, or environmental stresses (Kaushika and Rai 2007). The impact of different partial shading scenarios on PV module performance to track the PV module’s maximum power point was investigated by (Chalh et al. 2022). The P-V characteristic of the PV module is more complicated under partial shading due to the presence of some power peaks compared to uniform irradiance conditions where only one maximum power point exists. (Li et al. 2019) provided new insights into the comprehensive energy and economic performances of Photovoltaic Shading Systems (PVSS) in multi-story buildings. A numerical shading model was developed to evaluate the shading effect from an upper PVSS row on its subjacent row. The article (Nayan Kumar and Dey 2020) proposed a novel hybrid MPPT approach able to track the maximum possible power under any weather fluctuation, with comprehensive enhancement on all aspects of high performance, boosting the PV array efficiency by including a boost converter for controlling DC to DC power. The shortcomings of these algorithms are illustrated, and the analysis confirms the effectiveness of the proposed algorithm accordingly. Partial shading is one of the phenomena that lower the PV’s performance by creating hotspots (Sarwar et al. 2022) which are defined as local areas of elevated temperature. These hotspots show up as a result of long periods of partial shading occurrence on a specific area of the PV’s surface or due to the PV panels’ manufacturing defects.

The work of (Lin et al. 2020) aimed to provide PV module selection (thin film, monocrystalline, or polycrystalline) with better performance in the shading environment to improve the system’s efficiency. (Koondhar et al. 2021) explored design scenarios to increase the peak power when partial shading and mismatch problems occur, and provided an in-depth understanding for the researchers engaged in the field of PV systems and industries stepping up the production of PV cells. (Uno and Kukita 2015) showed various kinds of differential power processing converters and voltage equalisers for PV modules to prevent negative influences of partial shading. The impact of high PV cell temperature due to hot spotting was thoroughly studied (Dhimish et al. 2018). It showed that elevated temperatures could permanently damage the PV module. This highlights the importance of hot spot mitigation techniques. The impact of the shading effects caused by various obstructions, including trees, rooftop structures, and neighbouring buildings on the residential rooftop systems, was studied by (Bulanyi and Zhang 2014), where different levels of shading may result in PV system performance degradation.

Solar energy resources are getting more consideration as they promote the concept of sustainable living. During the past few decades, huge research has been conducted on the applications of bypass diodes (BPDs), buck-boost converters (Kermadi et al. 2020), and micro-inverters to improve the efficiency of solar modules by mitigating the impact of partial shading and hotspots. (Tsai 2010) implemented and analysed a novel model of PV module using Matlab/Simulink software package. They developed a model to tackle the V modules’ parameters as a function of the ambient conditions. Bypass diodes are included in PV modules to counter the hotspot formation that significantly affects PV modules’ performance in the long term (Savitha, Shashikala, and Puttabuddhi 2014). During the shading of a PV module, a mismatch in the whole PV system’s output current occurs. In the worst-case scenario, the current generated by the shaded module is zero (Pannebakker, Waal, and Sark 2017). This leads the shaded module to behave as a load and operate in the reverse-biased mode. Therefore, the voltage generated from the unshaded cells will push the shaded cell to the breakdown region, which will, in turn, lead the current to flow in a reverse direction, feeding into the shaded cell, causing it to heat up hence the rise of hotspots.

1.3. Contribution

The proposed system that allows PV modules to operate in different modes of the internally-connected PV submodules provides several options for controlling the output voltage and current and mitigating the effect of partial shading. The main contributions of this work are:

• It expands the connection type of the PV submodules to be in parallel or in series via an external circuit, including a set of switches

• Therefore, the output current and voltage can be varied without connecting extra PV modules in parallel or the need for a buck-boost converter.

• The effect of partial shading is minimised, and the hotspot effect is mitigated without bypass diodes.

• A Matlab/Simulink model is developed to validate the configurations, and two customised hardware PV modules, representing the conventional module and the proposed module, are assembled from the scratch.

• The implemented hardware and the results prove the concept of separating shaded PV submodules and controlling the internal connections of the submodules in different configurations.

1.4. Organization of the paper

This paper is organised as follows: Section 2 addresses the background information, Section 3 introduces the main method, section 4 discusses the results of experimental and simulation testing, and section 5 includes the main conclusions of this paper.

2. Background information

To investigate the effect of partial shading on PV performance, a PV system of four 120-W modules, as shown in Figure 1, is implemented in SIMULINK, and the shading percentage is increased from 0% to 75% as shown in Table 1.

Figure 1. Four series-connected 120W PV modules (a) without and (b) with BPDs.

The current and voltage of four series-connected photovoltaic modules are measured. The Figure has two parts; the upper four photovoltaic modules do not have bypass diodes. The lower four photovoltaic modules have bypass diodes. The rating of each photovoltaic module is 120 Watts.

Table 1. Shading percentages with and without BPDs.

Shaded Modules	Shading Percentage for Each Module	Without BPDs – Pic 5–5	With BPDs
None	0%	Case A	Case AA
Module 1	25%	Case B	Case BB
Modules 1 & 2	50%	Case C	Case CC
Modules 1& 2 & 3	75%	Case D	Case DD

The impact of partial shading on PV modules can be monitored by analysing their current-voltage (I-V) curves, as shown in Figure 2, at different irradiance levels, with and without BPDs. The PV modules are subjected to four levels of shading represented by cases (Case A-to-Case D) without BPD and by cases (Case AA-to-Case DD) with BPD. In case A and case AA, the PV modules are fully illuminated (0% shading). In case B and case BB, one PV module is shaded by 25%, while the remaining three modules are fully illuminated. In case C and case CC, two PV modules are shaded by 50%, while the other two modules are fully illuminated. In the last cases, case D and case DD, three PV modules are shaded by 75%, while the last module is fully illuminated. The corresponding I-V curves for the aforementioned cases are depicted in Figure 2, which compares the reference I-V curves (without BPDs) to the curves of the PV modules with BPDs. The drop in the maximum power point from case A/case AA to case D/case DD, respectively, is attributed to the drop in the short circuit current. It is clear in Figure 2 (b) that the inclusion of BPDs saves much power under fully shaded PV module conditions.

Figure 2. Current-voltage (I-V) curves of the PV system of Figure 1 (a) without and (b) with BPDs.

The figure has two subplots; the left subplot depicts four curves for the current-voltage characteristics of photovoltaic systems without bypass diodes, while the right subplot shows the same curves with bypass diodes. In both plots, the upper curve represents the photovoltaic modules without shading. The second curve represents the photovoltaic modules with 25 percent shading. The third curve represents the photovoltaic modules with 50 percent shading. The last curve (solid line) represents the photovoltaic modules with 75 percent shading.

3. Method

This article develops an algorithm to control the internal connections of the PV submodules automatically in parallel or in series to vary the output current and voltage, respectively. This work improves the performance of partially shaded PV modules without using bypass diodes. Electronic circuitry is developed to control the internal connections of the PV submodules in series, parallel, or a mix of series and parallel submodules based on demand. The algorithm is written in MATLAB and executes the PV model shown in Figure 3, where the switches ‘ON/OFF’ either connect or disconnect the PV submodules based on the value of the pulse applied. These pulses are controlled by a microcontroller in the experimental part of this work. Each PV submodule comprises nine series-connected cells.

Figure 3. Proposed model of the PV module. The switches are assigned to different names and disconnect or connect the PV submodules based on pulses received by the microcontroller unit (or by a MATLAB code in the simulation). Each submodule comprises nine series-connected cells.

The four Photovoltaic submodules are represented by submodule 1 to submodule 4. The switches are presented and controlled by MATLAB built-in pulse functions. The radiation level is 1000 W/m2. The current and the voltage of the whole photovoltaic system are tracked for different scenarios.

The PV modules are accompanied by a couple of voltage and current sensors to monitor and then transfer the voltage and the current readings, as shown in Figure 4, to the ATmega328 (an 8-bit, 28-Pin AVR Microcontroller), which compares the actual demand (i.e. desired load voltage and current) with the monitored readings.

Figure 4. Structure of the proposed system. The actuators in the controlling unit control the mode of the PV submodules’ connection.

The controlling unit in the upper left box is the central processing unit that controls the actuators (shown in the adjacent box) to switch on/off the internal connections of the PV submodules. This controller unit processes the desired and measured voltage and current readings (shown in the lower box) to choose the most suitable connections for the PV submodules.

Figure 4 shows the working principle of the proposed system. The controlling unit (STM32 is used in this work) is the central processing unit that controls the actuators to switch on/off the internal connections of the PV submodules. This controller unit is placed inside the junction box of the PV module and processes the desired and measured voltage and current readings to choose the most suitable connections for the PV submodules.

The controlling unit either isolates the shaded submodule(s) or varies the internal connections based on the current sensors’ readings or based on the desired demand. The actuators work as electrical switches to keep controlling the internal PV submodules’ based on the demand. The internal submodules’ can have different connections to meet the desired load ratings or to isolate the shaded submodule from the unshaded submodules.

3.1. Hardware design

In this work, four voltage and current sensors are used for the tested PV module, which comprises four submodules in order to monitor any possible shaded submodule. The generated current of the PV cells is linearly proportional to the irradiance. Therefore, the shaded submodule generates a lower current compared to the other unshaded submodules, and it should be separated from the unshaded submodules by controlling the internal connections of these submodules. The isolation of the shaded submodule avoids the occurrence of hotspots and helps prevent the entire PV module from any possible damage.

The hardware design circuit, as shown in Figure 5, illustrates the printed circuit board (PCB) developed to minimise the area needed for circuits’ components and their connections. The designed PCB performs the same functionality as the experimental design, with a smaller size, better protection schemes, and improved efficiency.

Figure 5. (A) Designed and (b) assembled versions of the manufactured printed circuit board for the proposed system.

The upper circuit illustrates the printed circuit board (PCB) developed to minimize the area needed for circuits’ components and their connections. The lower circuit illustrates the actual assembled printed circuit board. The designed PCB performs the same functionality as the experimental design with improved efficiency.

3.2. Modes of operation

The PV submodules are connected in three different modes: series, parallel, or mixed, as depicted in Table 2. In the series mode, all the 36-cells are connected in series; therefore, the current is lowest since the total PV current is dominated by the lowest generated current by any cell, while the voltage is maximum since the voltage builds up in the series-connection mode. When it comes to parallel mode, the four PV submodules are connected in parallel, and the current is highest while the voltage is minimum since the total PV module’s voltage is dominated by the lowest submodule’s voltage. In the mixed mode, 18-series connected cells are connected in parallel with the other 18-series connected cells resulting in moderate values of voltage and current. The modes of operation are selected based on the sensors’ readings and the demand. In the proposed design, the main switching mechanism is accomplished by the relays, where each relay is connected to a certain part of the PV submodules, allowing the switching between the three modes through a signal that is initiated by the microcontroller.

Table 2. Modes of Operation.

Mode	Ns (Series-Connected)	Np (Parallel-Connected)	Current	Voltage
Series	36	1	Lowest	Highest
Parallel	9	4	Highest	Lowest
Mixed	18	2	Moderate	Moderate

3.3. Assembled smart PV module

This subsection describes the assembled photovoltaic (PV) module that consists of 36 polycrystalline PV cells and an integrated printed circuit board (PCB) that controls the internal connections of four PV submodules. The PV submodules are connected to the PCB’s terminals and can be switched on or off using electronic switches represented by transistors. The microcontroller unit sends pulses to these switches to connect or disconnect the submodules based on the required mode of operation specified in Table 3. Figure 6 shows the schematic of the internal connections of the submodules.

Figure 6. Assembled smart PV module as a) illustration and b) real system. The internal connections of the PV submodules are connected to the terminals of the PCB. The switches (transistors) receive pulses from the microcontroller to connect/disconnect the submodules as depicted in Table 3. The switches of the same notation are shown multiple of times just for illustration.

Thirty-six polycrystalline photovoltaic cells with blue color and a PCB with green color are integrated into the same module to control the internal connections of the PV submodules. These connections are represented by transistors.

Table 3. Modes of operation and the corresponding switches’ states.

Mode	Closed switches
(Series) Four submodules are connected in series	S1 and S4
(Parallel) Four submodules are connected in parallel	S1 and S2
(Mixed) Two parallel-connected submodules are connected in series with the other two parallel-connected submodules.	S2, S3, and S4

4. Results

In this work, two PV modules are assembled comprising an identical number of the same polycrystalline PV cells; the first module mimics the conventional PV modules where all the cells’ connections are fixed. The second module (Smart PV module, as in Figure 6) has an extra feature where the internal connections of the PV submodules are variable and controlled by electronic switches. The different modes (as depicted in Table 2) are tested and compared with the conventional module.

4.1. Simulation results

The simulation is conducted in Matlab/Simulink, and results are categorised into the:

• Conventional module;

• Smart module: series-connected mode;

• Smart module: parallel-connected mode;

The results of the smart module for the series-connected mode are the same as those of the conventional PV modules. The smart PV module is tested for each mode with (one PV submodule is fully shaded) and without shading conditions. Table 4 illustrates the electrical parameters of the conventional PV submodules, where each submodule comprises nine series-connected cells. The proposed idea is tested using MATLAB under the following conditions: the solar irradiance is 1000 W/m² during no shading conditions while it is 0 W/m² throughout the shading, and the temperature is 25 Degree Celsius.

Table 4. PV submodule’s electrical parameters.

Parameter	Description	Value	Unit
Voc	Open-circuit voltage	5.8	V
Isc	Short-circuit current	9.8	A
Pmax	Maximum power	43.5	W
G	Solar irradiance	1000	W/m^2
Tc	Temperature	25	Degree Celsius

4.1.1. Conventional PV module (Smart module: series-connected mode)

The conventional and the smart PV modules (for the series-connected mode) are similar, where their current-voltage (I-V) and power-voltage (P-V) characteristics are tested under ambient conditions and shown in Figure 7 The PV modules’ open-circuit voltage V_ocM equals 23.2 V (5.8 × 4 = 23.2 V), the modules’ short-circuit current I_scM equals 9.8 A, and the modules’ maximum power PM equals 173.8W (43.5 × 4 = 174 W).

Figure 7. Current- and power-voltage characteristics of the conventional PV module without shading.

The figure has two subplots; the left subplot represents the current-voltage characteristics of the conventional photovoltaic module without shading, while the right subplot represents the power-voltage characteristics of the same conventional photovoltaic module without shading. The graphs indicate some points to show the open-circuit voltage, short-circuit current, and maximum power point values.

The module under the shading conditions is tested, and the characteristics of the series-connected mode with full shading on one of the PV submodules are demonstrated in Figure 8. The shaded PV submodule behaves as a reverse-biased diode consuming power; hence the overall current is the minimum short-circuit current of the connected submodules. Moreover, the module’s voltage drops to zero. As shown in Figure 8, the maximum current is 7.34 A and then drops dramatically to zero, while both the voltage and power are almost zero due to the shading. The bottom two plots of Figure 8 are similar to that of Figure 7, which proves the claim that the conventional and the smart PV modules (for the series-connected mode) match each other in the absence of the shades. However, during the shading, the bypass diode in the conventional module isolates two submodules while our proposed series-connected mode isolates the shaded submodule only.

Figure 8. Current- and power-voltage characteristics of the smart PV module (series-connected mode) with fully shading one submodule (the upper two plots) and without shading the submodules (the lower two plots).

The figure has four subplots for the series-connected submodules; the upper-left subplot represents the current-voltage characteristics of the smart photovoltaic module with shading, while the upper-right subplot represents the power-voltage characteristics of the same photovoltaic module with shading. The lower-left subplot represents the current-voltage characteristics of the smart photovoltaic module without shading, while the lower-right subplot represents the power-voltage characteristics of the same photovoltaic module without shading. The graphs indicate some points to show the open-circuit voltage, short-circuit current, and maximum power point values.

4.1.2. Parallel-connected mode

This section starts with the analysis of the parallel-connected submodules without shading conditions. Figure 9 shows the measured current- and power-voltage characteristics of the smart PV module for the parallel-connected mode with fully shading one submodule (the upper two plots) and without shading the submodules (the lower two plots). The results show that the short-circuit current (I_scM = 39.1 A) of the smart PV module is four times the short-circuit current (I_sc = 9.8 A) of each submodule, while the module’s open-circuit voltage V_ocM is around 5.8 V, which equals the open-circuit voltage Voc of each submodule. Therefore, the total power (P_M = 173 W) equals four times the power of each submodule (P_max = 43.3 W). Under shading conditions, one PV submodule of the parallel-connected PV module is fully shaded, the overall current of the module drops from 39.1 A to 29.3 A, and the voltage remains almost the same. This proves that the shaded submodule does not contribute to the overall current. The maximum power of the module is approximately P_M = 130 W which is 43W less than that of the no shading conditions.

Figure 9. Current- and power-voltage characteristics of the smart PV module (parallel-connected mode) with fully shading one submodule (the upper two plots) and without shading the submodules (the lower two plots).

The figure has four subplots for the parallel-connected submodules; the upper-left subplot represents the current-voltage characteristics of the smart photovoltaic module with shading, while the upper-right subplot represents the power-voltage characteristics of the same photovoltaic module with shading. The lower-left subplot represents the current-voltage characteristics of the smart photovoltaic module without shading, while the lower-right subplot represents the power-voltage characteristics of the same photovoltaic module without shading. The graphs indicate some points to show the open-circuit voltage, short-circuit current, and maximum power point values.

4.1.3. Mixed-connected mode

Similarly, this section starts with analysing the mixed-connected mode without shading conditions. Figure 10 shows the measured current- and power-voltage characteristics of the smart PV module for the mixed-connected mode with fully shading one submodule (the upper two plots) and without shading the submodules (the lower two plots). The results show that the short-circuit current (I_scM = 19.5 A) of the smart PV module is twice the short-circuit current (I_sc = 9.7 A) of each submodule, and the open-circuit voltage V_ocM is around 11.6 V, which equals twice the open-circuit voltage (V_oc = 5.8 V) of each submodule. Therefore, the total power (P_M = 173 W) equals four times the power of each submodule (P_max = 43.3 W). Under shading conditions, only one PV submodule of the mixed-connected PV module is fully shaded; the I-V and P-V curves of the smart PV module are depicted in the upper two plots of Figure 10. Since the fully-shaded submodule doesn’t contribute to the overall PV module’s current, the module still generates a current of 9.8 A. However, as the voltage is not affected significantly by shading submodules, the overall voltage is 11.4 V which is almost double the open-circuit voltage of one submodule. The total power, for this case, equals approximately 87.1 W, which is half the power for the unshaded conditions.

Figure 10. Current- and power-voltage characteristics of the smart PV module (mixed-connected mode) with fully shading one submodule (the upper two plots) and without shading the submodules (the lower two plots).

The figure has four subplots for the mixed-connected submodules; the upper-left subplot represents the current-voltage characteristics of the smart photovoltaic module with shading, while the upper-right subplot represents the power-voltage characteristics of the same photovoltaic module with shading. The lower-left subplot represents the current-voltage characteristics of the smart photovoltaic module without shading. In contrast, the lower-right subplot represents the power-voltage characteristics of the same photovoltaic module without shading. The graphs indicate some points to show the open-circuit voltage, short-circuit current, and maximum power point values.

4.2. Hardware implementation results

The PV module is tested under the following conditions: the inclination angle is 22.4°, the solar irradiance is 860–880 W/m², the ambient temperature is 41 ◦C, and the module’s temperature is 59 ◦C.

4.2.1. Conventional PV module

The PV analyser is used to measure the I-V curves and solar irradiance. The measurements are conducted on the assembled PV module and include the following three scenarios: without shading, with the shading of one cell, and with the shading of two cells. The PV module is exposed completely to the sun, and its electrical parameters are monitored by the PV analyser, as depicted in Table 5. The shading percentage is 75% for any shaded cell. The I-V curves of the test conventional PV module are measured without shading or with the shading of one cell in one of the submodules. The module includes two bypass diodes, one for every 18 cells (i.e. for two submodules). Without the presence of bypass diodes, the short-circuit current is low, and the power produced by this module is low. In the case of using bypass diodes, the corresponding two PV submodules (which include the shaded cell) are separated. Table 5 shows that the produced power of the conventional module is reduced from 12.75 W to 7.03 W.

Table 5. Conventional PV module’s parameters with and without shading conditions.

Parameter	Description	Measured Values			Unit
		Without shading	Shading one cell	Shading two cells
Voc	Open-circuit voltage	19.72	18.72	4.22	V
Isc	Short-circuit current	1.65	1.68	0.43	A
Pmax	Maximum power	12.75	7.03	18.90	W
Vmax	Voltage at maximum power	11.62	5.88	14.25	V
Imax	Current at maximum power	1.10	1.20	0.29	A

Table 5 also depicts the electrical parameters of the test PV module where one cell in the first submodule and another cell in the third submodule are shaded by 75%. The overall current decreases significantly as the current has a linear relationship with the irradiance. The measurements show that the short circuit current is reduced from 1.65 A to approximately 0.43 A.

4.2.2. Smart PV module

In this work, the PV submodules’ connections in this smart PV module are controlled by a microcontroller unit to get series-, parallel-, or mixed- connections). Any shaded PV submodule is automatically separated without the need for any bypass diode. The three cases are explained as follows:

4.2.2.1. Series-connected mode

The cells in the smart PV module are series-connected and exposed uniformly to solar irradiance without any shading for the cells. The PV analyser, I-V curve tracer, and wireless PV reference sensor are used to monitor the I-V characteristics of the PV module and the solar irradiance levels, respectively. Table 6 shows the electrical parameters of the proposed PV smart module without shading and with shading. The tested parameters are open-circuit voltage (V_oc), short-circuit current (I_sc), maximum power (P_max), voltage at maximum power (V_max), and current at maximum power (I_max).

Table 6. Comparison between the parameters of each of the three modes according to different shading conditions.

	Measured Values
	Without shading			Shading one cell			Shading two cells
Parameter	Ser.	Par.	Mix.	Ser.	Par.	Mix.	Ser.	Par.	Mix.	Unit
Voc	17.59	4.64	9.41	14.46	4.67	NA	9.67	4.67	NA	V
Isc	1.18	4.78	2.49	1.20	3.64	NA	1.17	2.32	NA	A
Pmax	9.84	8.31	9.60	7.67	6.25	NA	4.60	3.89	NA	W
Vmax	12.1	2.78	5.97	9.29	2.80	NA	5.98	2.82	NA	V
Imax	0.81	2.99	1.61	0.82	2.24	NA	0.76	1.43	NA	A

Under shading one cell conditions, the I-V curve of the smart PV module is measured where only one cell is shaded by 75% in one of the submodules. The module does not include bypass diodes (as of the conventional module). The open-circuit voltage decreases from 17.59 V to 14.46 V. The results prove that isolating the first PV submodule (the shaded submodule) mainly affects the voltage rather than the current of the series-connected cells. The results in Table 6 prove that the maximum power when one cell is 75% shaded is around 75% of the maximum produced power without shading.

When two cells are shaded, the I-V curve of the smart PV module is measured where one cell in the first submodule and another cell in the second submodule are shaded by 75%. The open-circuit voltage decreases from 17.59 V to 9.67 V. The results prove that isolating the first and the third PV submodules (the shaded submodules) mainly affects the voltage rather than the current of the series-connected cells. Table 6 also shows the characteristics of the smart PV module where the maximum power when two cells are 75% shaded is around 50% of the maximum produced power without shading.

4.2.2.2. Parallel-connected mode

The submodules in the smart PV module are parallel-connected and exposed to solar irradiance uniformly without any shading for the cells. The I-V characteristics of the PV module and the solar irradiance levels are monitored. Table 6 also shows the electrical parameters of the proposed PV smart module, where the total current is approximately four times the current of the series-connected module. On the other hand, the open-circuit voltage (Voc = 4.64 V) is a quarter of the open-circuit voltage of the series-connected mode (as depicted in Table 6).

Shading the PV cells by 75% is conducted when only one cell is shaded and then when two cells from different submodules are shaded. The results in Table 6 prove that when one cell in the first submodule is shaded, the sting is separated, and a reduction of 25% (from 4.78 V to 3.64 V) in the rated current results while the voltage is expected to remain constant. Therefore, the produced power, in this case, will be around 75% of the rated power. For the case when two cells are shaded with 75%, the electrical parameters are depicted in Table 6, where the first and the third submodules are shaded and automatically separated. Consequently, the current drops by 50% while the voltage is not impacted. Table 6 also shows that the current decreases from 4.78A to 2.32A with a negligible change in the voltage. The produced power, in this case, is reduced by nearly 50% compared to no shading conditions.

4.2.2.3. Mixed-connected mode

The smart PV module in the mixed-mode connection consists of two submodules connected in parallel, and each of these two submodules comprises eighteen series-connected cells. As a result, if any of the cells are shaded, the submodule corresponding to that cell will be isolated; consequently, two submodules will be in parallel to one submodule which will lead to voltage mismatch. Therefore, the mixed-mode connection will be used only without shading. If shading occurs, the connection will be changed to the parallel-connected mode or the series-connected mode based on the load and the conditions. The electrical properties are depicted in Table 6, obtained using the PV analyser, IV curve tracer, and the Wireless PV reference sensor. Since there are two PV submodules connected in parallel, the total current is twice the current of one submodule. Additionally, the voltage is twice the submodule’s voltage since each contains eighteen series cells.

5. Conclusion

To conclude, the proposed system that allows PV modules to operate in different modes of the internally-connected PV submodules provides several options for controlling the output voltage and current and mitigating the effect of partial shading. The system is initially modelled and implemented in Matlab/Simulink and is tested for different PV submodules’ configurations by controlling their internal connections. Then, the system is implemented practically where the PV submodules are controlled via electronic switches integrated with a designed PCB that comprises electronic switches to mitigate the effect of partial shading instead of using bypass diodes or buck-boost converters. Finally, a comparison between the hardware models of the conventional PV module and the proposed smart PV module was recorded. The results prove the practicality of the proposed system. In the future, further work will be proposed to adopt the system for large-scale projects.

Author contributions

Anas Al Tarabsheh came up with the concept for the project, carried out the simulation aspect, and authored the paper. The colleagues, Ahmed M. Abughali, Abdullah AlSalmani, and Ahmed J. AlSoufi, collaborated to construct the hardware as a team. Leen Ba’Ba’ played a role in finalising the manuscript for publication. The submitted version of the manuscript was reviewed and approved by all authors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

“This research was funded by the Office of Research and Sponsored Programs at Abu Dhabi University, grant number 19300608”

Notes on contributors

Anas Al Tarabsheh

Anas Al Tarabsheh received his B.Sc. and M.Sc. degrees in Electrical Engineering from Jordan University of Science and Technology and a Ph.D. degree in Electrical Engineering from Stuttgart University. He taught more than ten undergraduate courses in the Electrical Engineering department at Hashemite University from 2007 to 2014; then, He joined Abu Dhabi University in 2015 as an associate professor at the Electrical, Computer, and Biomedical Engineering department for the undergraduate and postgraduate programs. He received a couple of US patents and published many peer-reviewed articles in Photovoltaics. He is honored to hold three awards: the University Ambassador Award, the University Teaching Excellence Award, and the College of Engineering Teaching Excellence Award. His research interests are in Semiconductors and Renewable Energy.

Ahmed M. Abughali

Ahmed M. Abughali received his B.Sc. degree in Electrical Engineering from Abu Dhabi University. He is perusing his M.Sc. in Electrical Engineering at Khalifa University. His research interests are Renewable Energy, Space Systems, and Technology.

Abdullah M. AlSalmani

Abdullah M. AlSalmani received his B.Sc. degree in Electrical Engineering from Abu Dhabi University. He is perusing his M.Sc. in the National Space Science and Technology Center at United Arab Emirates University. His research interests are Renewable energy and aerospace.

Ahmed J. AlSoufi

Ahmed J. AlSoufi received his B.Sc. degree in Electrical Engineering from Abu Dhabi University. His research interests are Renewable energy and Electronics.

Leen B. Ba`ba`

Leen B. Ba`ba` received his B.Sc. degree in Electrical Engineering from Abu Dhabi University. She is currently a research assistant at Abu Dhabi University. Her research interests are AI, Electronics, and Renewable energy.