Defect analysis and performance evaluation of photovoltaic modules using quantitative electroluminescence imaging

photovoltaic power plant, performance model, electroluminescence

1. Introduction

The long-term performance of photovoltaic (PV) modules declines over time, influenced by environmental conditions such as temperature, humidity, and shading, which pose operational challenges. Quantifying this long-term degradation is crucial for predicting the return on investment of PV systems. Many studies have examined the degradation of both conventional crystalline silicon and thin-film PV technologies under real-world conditions, with reported degradation rates varying across technologies and climates.

The research by Som [1] found that monocrystalline and polycrystalline silicon panels have an average degradation rate of about 0.5% per year in temperate climates. However, Radue et al. [2] observed that thin-film modules, particularly a-Si, degrade faster, with rates reaching up to 1.5% annually, mainly due to exposure to light and high temperatures. Similarly, Silvestre et al. [3] analyzed four thin-film technologies and noted that CdTe degraded the fastest, while cadmium indium selenium remained the most stable. Smith et al. [4] also reported an annual power loss of 0.5% in crystalline silicon modules, mainly due to current decay. Rajput et al. [5], through a long-term analysis of multicrystalline modules over 22 years, calculated a degradation rate of 1.9% per year. Additionally, Faye [6] found that PV modules degrade through several mechanisms, including delamination (corrosion), hot spots, soiling, and cell cracks.

Moreover, Singh et al. [7] conducted a 12-year degradation and reliability analysis, comparing the performance of PV modules and identifying significant differences depending on the technology used, emphasizing the importance of long-term reliability assessments. Abdallah et al. [8] investigated the degradation of PV module backsheet materials in desert climates, uncovering major reliability challenges in harsh conditions due to prolonged exposure to high temperatures and ultraviolet (UV) radiation. These findings are summarized in Table 1 to provide an overview of the current knowledge on PV module degradation.

Table 1.

The methodology to research the degradation of PV module.

Study	PV technology	Observation time	Methodology	Key findings
Som [1]	Mono-Si, Multi-Si	Not specified	Outdoor experimental testing	Two first-generation mono-Si panels completely failed, severe corrosion
Radue et al. [2]	Thin-film a-Si:H and CIGS	1 year	Outdoor deployment of 5 modules	a-Si:H modules initially degraded due to light-induced effects
Silvestre et al. [3]	CdTe, copper indium gallium selenide (CIGS)	4 years	Degradation rate analysis	CIS module least degraded
Smith et al. [4]	Mono-Si, Multi-Si	19–22 years	I-V curve analysis	Power loss <0.5%/year
Rajput et al. [5]	Multi-Si	22 years	Degradation analysis	Power degradation rate 1.9%/year
Faye et al. [6]	Polycrystalline	Few years	Degradation analysis Casamance (Senegal) and Cologne (Germany)	8.75%/year to 22.45%/year in Casamance and 0.8%/year to 1,4%/year in Cologne
Singh et al. [7]	Amorphous silicon (a-si), Heterojunction Intrinsic thin-layer (HIT), and multicrystalline (mc-si)	12 Years	Degradation analysis	The average degradation rates are found to be 1.24%/year for a-Si, 0.14%/year for HIT, and 1.50%/year for mc-Si.
Abdallah et al. [8]	Polyamide (PA) and two different polyethylene terephthalate (PET)	10 years	PV backsheet degradation in desert climates Observed at the outdoor test facility	Embrittlement of PA and PET backsheet results from prolonged UV exposure, temperature cycles, and humidity, causing cracks in the UV layer and degradation.

Study	PV technology	Observation time	Methodology	Key findings
Som [1]	Mono-Si, Multi-Si	Not specified	Outdoor experimental testing	Two first-generation mono-Si panels completely failed, severe corrosion
Radue et al. [2]	Thin-film a-Si:H and CIGS	1 year	Outdoor deployment of 5 modules	a-Si:H modules initially degraded due to light-induced effects
Silvestre et al. [3]	CdTe, copper indium gallium selenide (CIGS)	4 years	Degradation rate analysis	CIS module least degraded
Smith et al. [4]	Mono-Si, Multi-Si	19–22 years	I-V curve analysis	Power loss <0.5%/year
Rajput et al. [5]	Multi-Si	22 years	Degradation analysis	Power degradation rate 1.9%/year
Faye et al. [6]	Polycrystalline	Few years	Degradation analysis Casamance (Senegal) and Cologne (Germany)	8.75%/year to 22.45%/year in Casamance and 0.8%/year to 1,4%/year in Cologne
Singh et al. [7]	Amorphous silicon (a-si), Heterojunction Intrinsic thin-layer (HIT), and multicrystalline (mc-si)	12 Years	Degradation analysis	The average degradation rates are found to be 1.24%/year for a-Si, 0.14%/year for HIT, and 1.50%/year for mc-Si.
Abdallah et al. [8]	Polyamide (PA) and two different polyethylene terephthalate (PET)	10 years	PV backsheet degradation in desert climates Observed at the outdoor test facility	Embrittlement of PA and PET backsheet results from prolonged UV exposure, temperature cycles, and humidity, causing cracks in the UV layer and degradation.

Table 1.

The methodology to research the degradation of PV module.

Study	PV technology	Observation time	Methodology	Key findings
Som [1]	Mono-Si, Multi-Si	Not specified	Outdoor experimental testing	Two first-generation mono-Si panels completely failed, severe corrosion
Radue et al. [2]	Thin-film a-Si:H and CIGS	1 year	Outdoor deployment of 5 modules	a-Si:H modules initially degraded due to light-induced effects
Silvestre et al. [3]	CdTe, copper indium gallium selenide (CIGS)	4 years	Degradation rate analysis	CIS module least degraded
Smith et al. [4]	Mono-Si, Multi-Si	19–22 years	I-V curve analysis	Power loss <0.5%/year
Rajput et al. [5]	Multi-Si	22 years	Degradation analysis	Power degradation rate 1.9%/year
Faye et al. [6]	Polycrystalline	Few years	Degradation analysis Casamance (Senegal) and Cologne (Germany)	8.75%/year to 22.45%/year in Casamance and 0.8%/year to 1,4%/year in Cologne
Singh et al. [7]	Amorphous silicon (a-si), Heterojunction Intrinsic thin-layer (HIT), and multicrystalline (mc-si)	12 Years	Degradation analysis	The average degradation rates are found to be 1.24%/year for a-Si, 0.14%/year for HIT, and 1.50%/year for mc-Si.
Abdallah et al. [8]	Polyamide (PA) and two different polyethylene terephthalate (PET)	10 years	PV backsheet degradation in desert climates Observed at the outdoor test facility	Embrittlement of PA and PET backsheet results from prolonged UV exposure, temperature cycles, and humidity, causing cracks in the UV layer and degradation.

Study	PV technology	Observation time	Methodology	Key findings
Som [1]	Mono-Si, Multi-Si	Not specified	Outdoor experimental testing	Two first-generation mono-Si panels completely failed, severe corrosion
Radue et al. [2]	Thin-film a-Si:H and CIGS	1 year	Outdoor deployment of 5 modules	a-Si:H modules initially degraded due to light-induced effects
Silvestre et al. [3]	CdTe, copper indium gallium selenide (CIGS)	4 years	Degradation rate analysis	CIS module least degraded
Smith et al. [4]	Mono-Si, Multi-Si	19–22 years	I-V curve analysis	Power loss <0.5%/year
Rajput et al. [5]	Multi-Si	22 years	Degradation analysis	Power degradation rate 1.9%/year
Faye et al. [6]	Polycrystalline	Few years	Degradation analysis Casamance (Senegal) and Cologne (Germany)	8.75%/year to 22.45%/year in Casamance and 0.8%/year to 1,4%/year in Cologne
Singh et al. [7]	Amorphous silicon (a-si), Heterojunction Intrinsic thin-layer (HIT), and multicrystalline (mc-si)	12 Years	Degradation analysis	The average degradation rates are found to be 1.24%/year for a-Si, 0.14%/year for HIT, and 1.50%/year for mc-Si.
Abdallah et al. [8]	Polyamide (PA) and two different polyethylene terephthalate (PET)	10 years	PV backsheet degradation in desert climates Observed at the outdoor test facility	Embrittlement of PA and PET backsheet results from prolonged UV exposure, temperature cycles, and humidity, causing cracks in the UV layer and degradation.

Despite these insights, while previous studies provide insights into PV degradation under temperate conditions, research focused on tropical monsoon climates, characterized by high humidity and temperatures, remains limited. This study aims to fill that gap by focusing on PV systems operating in these specific environments, which are characterized by high humidity and ambient temperatures ranging from 20°C to 35°C. This study addresses this gap by focusing on PV systems operating in tropical monsoon environments, with ambient temperatures ranging from 20°C to 35°C and an average humidity of 63%. The results indicate that thin-film and polycrystalline panels exhibit significantly higher degradation rates compared to monocrystalline technology, consistent with the findings of earlier studies. However, the specific impact of environmental factors, such as high humidity, requires further investigation.

Currently, three main technologies are used to detect defects in PV cells: electroluminescence (EL), infrared thermography (IRT), and photoluminescence (PL). EL is a method that applies electrical current to stimulate PV cells to emit light, thereby identifying defects such as cracks and performance degradation. This technique is particularly effective in detecting internal defects not visible to the naked eye [9]. IRT identifies thermal radiation emitted from PV modules to detect hotspots responsible for performance degradation. A key advantage of IRT is its ability to conduct outdoor inspections without interrupting system operations; however, it is less effective at identifying small defects that do not generate heat [10]. PL measures the light emitted from PV cells when illuminated by a strong light source, making it useful for detecting deep material defects, though it is more expensive and less effective at detecting thermal-related issues [11].

To further understand how weather impacts PV module degradation, this study also explores the use of EL imaging, which has become an effective technique for defect detection and performance evaluation of PV modules during both manufacturing and field operation. As shown in Table 2, EL imaging has proven valuable for identifying defects and assessing PV module performance. Seigneur et al. [12] discuss the capability of EL testing to detect the dynamic opening and closure of cracks within modules. Lai et al. [13] highlight the use of specialized digital image processing techniques that enable rapid analysis of EL data, replacing subjective visual interpretation. A major benefit, demonstrated across multiple studies, is the ability to correlate EL imaging metrics with electrical parameters such as power loss and fill factor reduction. For instance, Bedrich et al. [14] utilize quantitative EL analysis to estimate power losses resulting from potential-induced degradation. Similarly, Bazzoli et al. [15] and Puranik et al. [16] have developed generalized EL quantification techniques to assess module power output with low error margins.

Table 2.

The comparison of different methods using EL imaging to analyze the degradation of PV modules.

Study	Methodology	Key findings
Lai et al. [9]	Quantitative EL image analysis	2.5% lower EL intensity correlated to 0.35% efficiency loss
Seigneur et al. [12]	EL measurement before IV	Current/temperature variations cause arbitrary crack open/closure
Lai et al. [13]	EL imaging and digital analysis	Effective technique to assess PV degradation
Bedrich et al. [14]	Quantitative EL analysis	Estimate power loss due to potential-induced degradation
Bazzoli et al. [15]	Low-cost EL imaging for PV	Established EL imaging technique for PV
Puranik et al. [16]	Generalized quantitative EL method	Evaluate module power with 3% error
This study	Quantitative EL image analysis	Degradation rates of different PV technologies after few years of operation and common defects

Study	Methodology	Key findings
Lai et al. [9]	Quantitative EL image analysis	2.5% lower EL intensity correlated to 0.35% efficiency loss
Seigneur et al. [12]	EL measurement before IV	Current/temperature variations cause arbitrary crack open/closure
Lai et al. [13]	EL imaging and digital analysis	Effective technique to assess PV degradation
Bedrich et al. [14]	Quantitative EL analysis	Estimate power loss due to potential-induced degradation
Bazzoli et al. [15]	Low-cost EL imaging for PV	Established EL imaging technique for PV
Puranik et al. [16]	Generalized quantitative EL method	Evaluate module power with 3% error
This study	Quantitative EL image analysis	Degradation rates of different PV technologies after few years of operation and common defects

Table 2.

The comparison of different methods using EL imaging to analyze the degradation of PV modules.

Study	Methodology	Key findings
Lai et al. [9]	Quantitative EL image analysis	2.5% lower EL intensity correlated to 0.35% efficiency loss
Seigneur et al. [12]	EL measurement before IV	Current/temperature variations cause arbitrary crack open/closure
Lai et al. [13]	EL imaging and digital analysis	Effective technique to assess PV degradation
Bedrich et al. [14]	Quantitative EL analysis	Estimate power loss due to potential-induced degradation
Bazzoli et al. [15]	Low-cost EL imaging for PV	Established EL imaging technique for PV
Puranik et al. [16]	Generalized quantitative EL method	Evaluate module power with 3% error
This study	Quantitative EL image analysis	Degradation rates of different PV technologies after few years of operation and common defects

Study	Methodology	Key findings
Lai et al. [9]	Quantitative EL image analysis	2.5% lower EL intensity correlated to 0.35% efficiency loss
Seigneur et al. [12]	EL measurement before IV	Current/temperature variations cause arbitrary crack open/closure
Lai et al. [13]	EL imaging and digital analysis	Effective technique to assess PV degradation
Bedrich et al. [14]	Quantitative EL analysis	Estimate power loss due to potential-induced degradation
Bazzoli et al. [15]	Low-cost EL imaging for PV	Established EL imaging technique for PV
Puranik et al. [16]	Generalized quantitative EL method	Evaluate module power with 3% error
This study	Quantitative EL image analysis	Degradation rates of different PV technologies after few years of operation and common defects

Additionally, the deep learning method explored by Liu et al. [17, 18] offers a complementary approach by using convolutional neural networks to automatically detect and classify defects without the need for detailed physical modeling. This method excels in processing efficiency and accuracy for large datasets. However, its reliance on the diversity and quality of the training dataset may limit its ability to detect unusual or subtle defects, particularly in cases with uncommon defect patterns. In contrast, the quantitative EL method in this study requires no training data, offering greater flexibility and adaptability in various conditions. Furthermore, while the deep learning approach effectively detects surface defects, it may struggle with hidden issues such as small shunts and dark spots, which are accurately identified by the EL method used here. The EL method also provides precise quantitative analysis by correlating image intensity with degradation parameters over time, making it more suitable for contexts requiring detailed performance assessments.

While many studies have employed the EL method to assess the performance degradation of PV modules, prior research has largely focused on temperate climate conditions or analyzed individual PV technologies in isolation. This study makes a novel contribution by introducing a comprehensive quantitative EL approach to evaluate the performance and identify defects across three major PV technologies—thin-film, monocrystalline, and polycrystalline—under tropical monsoon climate conditions, an area that has not been thoroughly explored. Specifically, the study examines the degradation behavior of thin-film PV modules over 5 years, monocrystalline silicon modules over 3 years, and polycrystalline silicon modules over 12 years. The research is organized as follows: Section 2 describes the methodology, Section 3 outlines the experimental setup, Section 4 presents the results along with the discussion, and Section 5 provides the conclusion.

2. Methodology

The degradation of PV cells can be evaluated using various methods, such as the current intensity measurement method and EL imaging. In this study, the current intensity measurement method is employed to quantify the annual performance degradation, while the EL imaging method is utilized to detect and analyze defects during both the production and operational phases.

The degradation rate represents the decline in the performance of PV modules over time as a percentage of their initial performance. This rate is determined by assessing changes in key parameters, including maximum power output (P_max), fill factor (FF), open-circuit voltage (V_oc), and short-circuit current (I_sc). As a critical indicator, the degradation rate plays a foundational role in evaluating the long-term reliability and durability of PV technologies under various environmental conditions.

In this study, Equations (1)–(7) provide the physical and mathematical basis for describing the current–voltage (I-V) and power–voltage (P-V) characteristics of PV modules under both standard and real-world operating conditions. Building on this foundation, Equations (8)–(11) are applied to calculate the degradation rate. Details of the specific calculations are sequentially presented in the following sections.

2.1 One-diode model for solar cell

Based on the technical specifications provided by the manufacturer, the mathematical model of the solar cell was built on a single-diode model. The I-V and P-V output characteristic equation of PV device is given by [19]

\begin{array}{l} I = N_{P} I_{P H} - N_{P} I_{S} {\exp [\frac{q}{k T_{P V} A} (\frac{V}{N_{S}} + \frac{I R_{S}}{N_{P}})] - 1} - \frac{1}{R_{S H}} (\frac{N_{P} V}{N_{S}} + I R_{S}) \end{array}

(1)

where $N_{P}$ and $N_{S}$ are the numbers of solar cells in parallel and in series, $I_{P H}$ is the photocurrent, $I_{S}$ is the cell saturation of dark current, q is the charge of an electron, k is the Boltzmann’s constant, $T_{P V}$ is the cell temperature that is assumed to be uniform in the PV module, A is the ideal factor that depends on PV technology, $R_{S H}$ and $R_{S}$ are the resistance of shunt and series resistors, and $\exp (\cdot)$ is the exponential function. The photocurrent mainly depends on the solar irradiance and cell’s working temperature and is obtained by

\begin{array}{l} I_{P H} = [I_{S C}^{S T C} + K_{I} (T_{P V} - T_{P V}^{S T C})] \frac{G}{G^{S T C}} \end{array}

(2)

where $I_{S C}^{S T C}$ is the short-circuit current of PV module at the standard test condition of $T_{P V}^{S T C} = 25^{\circ} C$ and $G^{S T C} = 1 k W / m^{2}$ ⁠, $K_{I}$ is the temperature coefficient of cell’s short-circuit current. In addition, the cell’s saturation current varies with the cell temperature and is described as

\begin{array}{l} I_{S} = I_{R S} {(\frac{T_{P V}}{T_{P V}^{S T C}})}^{3} \exp [\frac{q E_{B G}}{k A} (\frac{1}{T_{P V}^{S T C}} - \frac{1}{T_{P V}})] \end{array}

(3)

where $I_{R S}$ is the cell’s reverse saturation current, $E_{B G}$ is the band-gap energy of the semiconductor used in the cell. Having an operating voltage, the PV output power is calculated by

\begin{array}{l} P_{P V} = I V \end{array}

(4)

2.2 EL measurement

PV modules often develop defects during manufacturing and operation, leading to power loss. While some defects can be visually inspected, accurately assessing defects requires precise measurement and modeling techniques. EL imaging is a highly effective technique used for identifying and analyzing defects in PV modules. By applying electrical current to the module, EL imaging induces light emission from PV cells, allowing visualization of micro-defects that are otherwise invisible to the naked eye. This method offers a high level of accuracy for identifying performance-limiting issues such as cracks, shunts, and hotspots. It involves increasing the direct current (DC) across the PV panel until the details of individual cells become apparent, while maintaining a constant voltage equivalent to the panel’s V_oc. These researchers have identified and calculated various defects in monocrystalline and polycrystalline PV modules. Furthermore to calculate the EL current (⁠ $I_{E L}$ ⁠) and intensity of EL (⁠ $ϕ_{E L}$ ⁠) emission based on the one-diode model and forward voltage as per the following equations [11, 20–23].

\begin{array}{l} I_{E L} = I_{s} \exp (\frac{v_{f}}{n_{1} v_{t}}) \end{array}

(5)

\begin{array}{l} V_{t} = \frac{k t}{q} \end{array}

(6)

\begin{array}{l} ϕ_{E L} = A \exp (\frac{v_{f}}{v_{t}}) \end{array}

(7)

where A is the voltage calibration constant.

Additionally, to evaluate the statistically significant differences in degradation rates among PV technologies, a one-way analysis of variance (ANOVA) was conducted. Degradation rate data from three PV technologies (thin-film, monocrystalline, and polycrystalline) were used as the basis for comparison. Upon detecting significant differences, a Tukey honestly significant differences post-hoc test was performed to specifically identify the pairs of technologies with differences. The significance level was set at P < .05 for all statistical tests.

2.3 The effect of shunt and shading on the current flow of PV panel

In the field of solar energy, shading is a major factor contributing to the degradation of PV systems. Shading can occur due to two primary factors. First, it can be caused by external factors such as clouds, which abruptly reduce solar radiation and thereby diminish the efficiency of PV systems. Another factor contributing to shading is dead cells, which obstruct current flow and impact the performance of solar panels. This process typically occurs within PV panels during operation, influenced by weather conditions such as humidity and high temperatures. Research by Meyer et al. [24] indicates that shading reduces shunt resistance by approximately 30% when a module is shaded. Power losses due to shading amount to approximately 2.6% for 10% module area coverage and can increase to 38% for 50% coverage. A report by First Solar company [25] shows that sixth-generation PV modules experience shading losses ranging from 0.3% to 1.8% in the first year. Shunts, visible as “spider” defects on PV cells, divert current from its intended path, leading to permanent power and energy losses due to inefficient current flow. Specifically, power losses can reach approximately 7%–10% under full sunlight (1 sun) and up to 30%–40% under reduced sunlight conditions (0.2 suns), as illustrated in Fig. 1. To mitigate these issues, several strategies are employed in the field of solar energy. First, current flow is restricted to narrow pathways within PV cells to prevent unintended shunt paths. This approach reduces the severity of shunting occurrences and minimizes their impact on overall PV performance. Additionally, efforts are made to minimize current flow through shunt pathways during their formation, particularly in response to shading and other environmental factors. Furthermore, ongoing measures aim to reduce power and energy losses associated with existing shunts caused by previous shading events.

Figure 1.

Analysis of the effect of shunt and shading on current flow in PV panels [24].

2.4 The mathematical formula for degradation

To calculate the degradation of PV cells using the EL measurement method, compare the power under standard conditions published by the manufacturer at the time of shipment with the power at the time of measurement. Let $P_{r e f}$ be the peak power published by the manufacturer and $P_{E L}$ be the power at the time of measurement. The degradation of the PV cell is calculated using the following equations:

\begin{array}{l} P_{D} = P_{r e f} - P_{E L} \end{array}

(8)

\begin{array}{l} P_{D r} = \frac{P_{D}}{y_{r}} \end{array}

(9)

\begin{array}{l} D r = \frac{P_{D r} 100}{P_{r e f} y_{r}} \end{array}

(10)

And the FF can be calculated by the following equation:

\begin{array}{l} F F = \frac{P_{r e f}}{V_{O C} I_{S C}} \end{array}

(11)

where $P_{D r}$ (A) current intensity decreases annually, $y_{r}$ is the number of years of PV panel operation, $D r$ is annual degradation rate (%), $F F$ (%) is the fill factor, $V_{O C}$ is open-circuit voltage, $I_{S C}$ is a short-circuit current.

In this study, the S6 FS300 EL measurement system was used to measure the open-circuit voltage and to determine the number of inactive cells (dark cells). An increase in the number of dark cells within a cell can impede current flow, and when it reaches a certain threshold, it may lead to cell failure. Furthermore, through EL imaging, if multiple dark spots are observed at the connections between cells, this indicates that the connections are no longer conductive. The larger the area of the dark spots, the greater the reduction in conductivity at the connections. If the dark spots cover the entire connection area, it indicates that the connection between the cells has been completely lost.

2.5 Experimental setup

Figure 2 depicts the experimental setup for testing PV panels using the EL method. The system includes EL illumination source, six infrared cameras, DC power supply to energize the panels, high-definition multimedia interface display to configure input parameters and show I-V and P-V characteristics during measurement. To evaluate the performance and defects of the panels, seven steps are taken:

Figure 2.

Experimental setups for testing PV panels: (a) Structure of the EL measurement system; (b) process of measuring PV panels using the EL system.

Steps 1 and 2: PV panels under test are placed on racks and fed to the simulator to acquire I-V and P-V indexes. Next, the panel is positioned accurately on the machine, technical specifications are verified and tuned to match system requirements. The EL lamp is then turned on at 1000 W/m² intensity to excite the PV cells. Output DC current is collected and transmitted to the mainboard with integrated I-V and P-V analytics. After analysis, results are shown on the HMI screen, including I-V and P-V trends of the panel such as temperature, I_sc, V_oc and P_max.

Next, upon finishing the simulation measurement, each cell of the panel undergoes inspection using the EL machine. A predefined voltage and current level are applied to the panel. Concurrently, the six infrared cameras scan longitudinally across the surface for a complete view. The operator can configure different current levels up to four ranges. Each level yields an image reflecting the overall status of the cells, detecting dark or dead cells. Brightness depends on test duration, test type, and current level (higher current induces higher brightness).

Finally, after capturing images of the cell condition, power cables are disconnected from the panel. The robotic arm lifts the panel out and places it onto a prepared rack, before processing the next one.

In this study, common PV technologies, including thin-film, monocrystalline silicon, and polycrystalline silicon modules, were used for the experiment. The thin-film PV panels were installed on the roof of First Solar’s facilities located in Cu Chi, Ho Chi Minh City, Vietnam. Monocrystalline and polycrystalline silicon panels were also installed in Ho Chi Minh City, Vietnam. These panels are currently operating under normal conditions. The climate in this region is characterized by temperatures ranging from 25°C to 35°C and an average humidity of 63%. The technical specifications of the three PV cell types tested are presented in Table 3.

Table 3.

Specification of PV model for testing.

Electrical specifications	Thin-film	Polycrystalline PV module	Monocrystalline PV module
Module type	FS-6470A-P	RS-P618-60W	PT-110Wp	WE-110P-36
Nominal power (W)	470	60	110	110
Efficiency (%)	17–18.2	14.54	19	16,54
Voltage at P_max— $V_{m p}$ (V)	191.1	17.6	20.83	18
Current at P_max—I_mp (A)	2.46	3.41	5.28	6.11
Open-circuit voltage— $V_{o c}$ (V)	224.3	21.60	24.48	21.90
Short-circuit current— $I_{s c} (A)$	2.61	3.74	5.69	6.47
Size of module (mm)	2009×1232×49	794×518×35	1070×540×30	1100×670×35
Weight (kg)	36	5.5	6.2	7.4

Electrical specifications	Thin-film	Polycrystalline PV module	Monocrystalline PV module
Module type	FS-6470A-P	RS-P618-60W	PT-110Wp	WE-110P-36
Nominal power (W)	470	60	110	110
Efficiency (%)	17–18.2	14.54	19	16,54
Voltage at P_max— $V_{m p}$ (V)	191.1	17.6	20.83	18
Current at P_max—I_mp (A)	2.46	3.41	5.28	6.11
Open-circuit voltage— $V_{o c}$ (V)	224.3	21.60	24.48	21.90
Short-circuit current— $I_{s c} (A)$	2.61	3.74	5.69	6.47
Size of module (mm)	2009×1232×49	794×518×35	1070×540×30	1100×670×35
Weight (kg)	36	5.5	6.2	7.4

Table 3.

Specification of PV model for testing.

Electrical specifications	Thin-film	Polycrystalline PV module	Monocrystalline PV module
Module type	FS-6470A-P	RS-P618-60W	PT-110Wp	WE-110P-36
Nominal power (W)	470	60	110	110
Efficiency (%)	17–18.2	14.54	19	16,54
Voltage at P_max— $V_{m p}$ (V)	191.1	17.6	20.83	18
Current at P_max—I_mp (A)	2.46	3.41	5.28	6.11
Open-circuit voltage— $V_{o c}$ (V)	224.3	21.60	24.48	21.90
Short-circuit current— $I_{s c} (A)$	2.61	3.74	5.69	6.47
Size of module (mm)	2009×1232×49	794×518×35	1070×540×30	1100×670×35
Weight (kg)	36	5.5	6.2	7.4

Electrical specifications	Thin-film	Polycrystalline PV module	Monocrystalline PV module
Module type	FS-6470A-P	RS-P618-60W	PT-110Wp	WE-110P-36
Nominal power (W)	470	60	110	110
Efficiency (%)	17–18.2	14.54	19	16,54
Voltage at P_max— $V_{m p}$ (V)	191.1	17.6	20.83	18
Current at P_max—I_mp (A)	2.46	3.41	5.28	6.11
Open-circuit voltage— $V_{o c}$ (V)	224.3	21.60	24.48	21.90
Short-circuit current— $I_{s c} (A)$	2.61	3.74	5.69	6.47
Size of module (mm)	2009×1232×49	794×518×35	1070×540×30	1100×670×35
Weight (kg)	36	5.5	6.2	7.4

3. Results and discussion

3.1 Performance degradation analysis of Thin-Film PV panels

The EL imaging results of the five thin-film PV panels are presented in Table 4, including the main technical parameters after 5 years of operation and images showing the condition of the thin-film modules, obtained using EL imaging technology. To identify the dark cell points on the thin-film panels, the EL current was gradually increased in increments of 0.5 A until it reached 2.5 A. At this point, the dark cell points became clearly visible and could be observed with the naked eye. The results recorded the number of dark cells present in each module.

Table 4.

The EL results of five thin-film PV panels after 5 years of operation.

No.	EL image	Specification
1		$I_{s c}$ ⁠: 2.604 A $I_{m p}$ ⁠: 2.378 A $V_{m p}$ ⁠: 180.394 V $P_{m a x}$ ⁠: 428.910 W $V_{o c}$ ⁠: 225.47 V FF: 0.730% 21 POint dark cell
2		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 181.509 V $P_{m a x}$ ⁠: 431.522 W $V_{o c}$ ⁠: 227.135 V FF: 0.730% 40-point dark cell
3		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.371 A $V_{m p}$ ⁠: 180.287 V $P_{m a x}$ ⁠: 427.505 W $V_{o c}$ ⁠: 225.505 V FF: 0.728% 40-point dark cells
4		$I_{s c}$ ⁠: 2.605 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 180.652 V $P_{m a x}$ ⁠: 429.491 W $V_{o c}$ ⁠: 226.767 V FF: 0.727% 44-point dark cells
5		$I_{s c}$ ⁠: 2.595 A $I_{m p}$ ⁠: 2.243 A $V_{m p}$ ⁠: 168.216 V $P_{m a x}$ ⁠: 377.367 W $V_{o c}$ ⁠: 219.272 V FF: 0.663% 18-point dark cell

No.	EL image	Specification
1		$I_{s c}$ ⁠: 2.604 A $I_{m p}$ ⁠: 2.378 A $V_{m p}$ ⁠: 180.394 V $P_{m a x}$ ⁠: 428.910 W $V_{o c}$ ⁠: 225.47 V FF: 0.730% 21 POint dark cell
2		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 181.509 V $P_{m a x}$ ⁠: 431.522 W $V_{o c}$ ⁠: 227.135 V FF: 0.730% 40-point dark cell
3		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.371 A $V_{m p}$ ⁠: 180.287 V $P_{m a x}$ ⁠: 427.505 W $V_{o c}$ ⁠: 225.505 V FF: 0.728% 40-point dark cells
4		$I_{s c}$ ⁠: 2.605 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 180.652 V $P_{m a x}$ ⁠: 429.491 W $V_{o c}$ ⁠: 226.767 V FF: 0.727% 44-point dark cells
5		$I_{s c}$ ⁠: 2.595 A $I_{m p}$ ⁠: 2.243 A $V_{m p}$ ⁠: 168.216 V $P_{m a x}$ ⁠: 377.367 W $V_{o c}$ ⁠: 219.272 V FF: 0.663% 18-point dark cell

Table 4.

The EL results of five thin-film PV panels after 5 years of operation.

No.	EL image	Specification
1		$I_{s c}$ ⁠: 2.604 A $I_{m p}$ ⁠: 2.378 A $V_{m p}$ ⁠: 180.394 V $P_{m a x}$ ⁠: 428.910 W $V_{o c}$ ⁠: 225.47 V FF: 0.730% 21 POint dark cell
2		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 181.509 V $P_{m a x}$ ⁠: 431.522 W $V_{o c}$ ⁠: 227.135 V FF: 0.730% 40-point dark cell
3		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.371 A $V_{m p}$ ⁠: 180.287 V $P_{m a x}$ ⁠: 427.505 W $V_{o c}$ ⁠: 225.505 V FF: 0.728% 40-point dark cells
4		$I_{s c}$ ⁠: 2.605 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 180.652 V $P_{m a x}$ ⁠: 429.491 W $V_{o c}$ ⁠: 226.767 V FF: 0.727% 44-point dark cells
5		$I_{s c}$ ⁠: 2.595 A $I_{m p}$ ⁠: 2.243 A $V_{m p}$ ⁠: 168.216 V $P_{m a x}$ ⁠: 377.367 W $V_{o c}$ ⁠: 219.272 V FF: 0.663% 18-point dark cell

No.	EL image	Specification
1		$I_{s c}$ ⁠: 2.604 A $I_{m p}$ ⁠: 2.378 A $V_{m p}$ ⁠: 180.394 V $P_{m a x}$ ⁠: 428.910 W $V_{o c}$ ⁠: 225.47 V FF: 0.730% 21 POint dark cell
2		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 181.509 V $P_{m a x}$ ⁠: 431.522 W $V_{o c}$ ⁠: 227.135 V FF: 0.730% 40-point dark cell
3		$I_{s c}$ ⁠: 2.602 A $I_{m p}$ ⁠: 2.371 A $V_{m p}$ ⁠: 180.287 V $P_{m a x}$ ⁠: 427.505 W $V_{o c}$ ⁠: 225.505 V FF: 0.728% 40-point dark cells
4		$I_{s c}$ ⁠: 2.605 A $I_{m p}$ ⁠: 2.377 A $V_{m p}$ ⁠: 180.652 V $P_{m a x}$ ⁠: 429.491 W $V_{o c}$ ⁠: 226.767 V FF: 0.727% 44-point dark cells
5		$I_{s c}$ ⁠: 2.595 A $I_{m p}$ ⁠: 2.243 A $V_{m p}$ ⁠: 168.216 V $P_{m a x}$ ⁠: 377.367 W $V_{o c}$ ⁠: 219.272 V FF: 0.663% 18-point dark cell

To determine the degradation rate of PV panels over a period, Equation (8) was applied. Table 5 presents the degradation results of five thin-film PV panels after 5 years of operation. The data shows that, after 5 years of outdoor use, the power output of the panels is as follows: 428.910 W out of a total of 470 W, 429.491 W out of 470 W, 427.505 W out of 470 W, 431.522 W out of 470 W, and 377.367 W out of 470 W. This corresponds to power losses of 41.090 W, 40.509 W, 42.495 W, 38.478 W, and 92.633 W, respectively.

Table 5.

Degradation of thin-film panels after 5 years of operation polycrystalline PV panel.

Panel NO	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	470	428.910	41.09	8.218	1.48	0.80
No.2	470	429.491	40.51	8.102	1.46	0.79
No.3	470	427.505	42.50	8.5	1.53	0.80
No.4	470	431.522	38.48	7.696	1.39	0.79
No.5	470	377.367	92.63	18.526	3.35	0.82

Panel NO	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	470	428.910	41.09	8.218	1.48	0.80
No.2	470	429.491	40.51	8.102	1.46	0.79
No.3	470	427.505	42.50	8.5	1.53	0.80
No.4	470	431.522	38.48	7.696	1.39	0.79
No.5	470	377.367	92.63	18.526	3.35	0.82

Table 5.

Degradation of thin-film panels after 5 years of operation polycrystalline PV panel.

Panel NO	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	470	428.910	41.09	8.218	1.48	0.80
No.2	470	429.491	40.51	8.102	1.46	0.79
No.3	470	427.505	42.50	8.5	1.53	0.80
No.4	470	431.522	38.48	7.696	1.39	0.79
No.5	470	377.367	92.63	18.526	3.35	0.82

Panel NO	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	470	428.910	41.09	8.218	1.48	0.80
No.2	470	429.491	40.51	8.102	1.46	0.79
No.3	470	427.505	42.50	8.5	1.53	0.80
No.4	470	431.522	38.48	7.696	1.39	0.79
No.5	470	377.367	92.63	18.526	3.35	0.82

From the data presented in Table 5, it is evident that the degradation rate of thin-film panels stabilizes around 7%–8% after five years of operation. Panel No. 5 shows a higher degradation rate of 18.526%, likely due to environmental stress or potential structural flaws. This observation indicates that Panel No. 5 is particularly susceptible to environmental stressors, such as high humidity and fluctuating temperatures, or may have inherent structural vulnerabilities. These findings highlight the critical role environmental conditions play in accelerating PV module degradation the variations in the FF among the panels provide insights into their capacity to sustain photovoltaic conversion efficiency, which represents the ability of PV cells to convert incident sunlight into electrical power, even as degradation progresses

The degradation of PV panels is influenced by various factors, primarily environmental and structural. To identify the key contributors to this degradation, EL imaging was employed, with results detailed in Fig. 3. The analysis revealed the presence of dark cells, indicative of small shunt defects—imperfections undetectable by conventional inspection methods. As shown in Fig. 3a, these dark cells act as barriers to current flow, disrupting the internal current pathways and leading to the formation of hotspots, which degrade efficiency over time.

Figure 3.

Analysis results using EL imaging of thin-film PV panels. (a) Panel No.2; (b) Panel No.5.

Prolonged exposure to high ambient temperatures and humidity exacerbates these issues by causing material expansion and corrosion within the panels. In the study area, with ambient temperatures ranging between 20°C and 35°C and average humidity levels of approximately 63%, these conditions accelerate the formation of shunts. As illustrated in Fig. 3b, these shunts obstruct current flow and create shaded areas, significantly reducing current intensity and overall power output.

In more severe instances, when the shaded areas exceed 50 mm in size, the obstruction becomes critical, causing operational imbalances in PV cells. These larger shaded regions reduce the panel’s energy production capacity, create uneven heat distribution, and further accelerate degradation. EL imaging has proven effective in identifying these shaded areas, which are often caused by moisture accumulation, as key contributors to power loss. This highlights the importance of timely maintenance to address moisture-related defects and sustain panel efficiency.

Furthermore, as depicted in Table 4 and Fig. 3a, dark regions corresponding to shunt defects in thin-film modules are associated with significant power loss. The quantification of these defects reveals an annual degradation rate of approximately 7%–8%, as shown in Table 5. These findings underscore the critical relationship between defect size, degradation rates, and the utility of EL imaging as a diagnostic tool for evaluating PV panel performance and guiding maintenance strategies.

3.2 Performance degradation analysis of polycrystalline PV panels

For polycrystalline PV panels, performance degradation is often influenced by factors such as hotspots, micro-cracks, potential-induced degradation, delamination, and the presence of dark cells. The measurement results using EL technology on two PV panels after 12 years of operation are presented in Table 6.

Table 6.

The EL results of two polycrystalline PV panels after 12 years of operation.

Module No. 1		Module No. 2
EL image	Specification	EL image	Specification
	$I_{s c}$ ⁠: 3.2477 A $I_{m p}$ ⁠: 2.9572 A $V_{m p}$ ⁠: 17.5093 V $P_{m a x}$ ⁠: 51.7783 W $V_{o c}$ ⁠: 22.785 V FF: 0.699		$I_{s c}$ ⁠: 3.0633 A $I_{m p}$ ⁠: 2.8703 A $V_{m p}$ ⁠: 17.5056 V $P_{m a x}$ ⁠: 50.2469 W $V_{o c}$ ⁠: 22.3169 V FF: 0.735%

Module No. 1		Module No. 2
EL image	Specification	EL image	Specification
	$I_{s c}$ ⁠: 3.2477 A $I_{m p}$ ⁠: 2.9572 A $V_{m p}$ ⁠: 17.5093 V $P_{m a x}$ ⁠: 51.7783 W $V_{o c}$ ⁠: 22.785 V FF: 0.699		$I_{s c}$ ⁠: 3.0633 A $I_{m p}$ ⁠: 2.8703 A $V_{m p}$ ⁠: 17.5056 V $P_{m a x}$ ⁠: 50.2469 W $V_{o c}$ ⁠: 22.3169 V FF: 0.735%

Table 6.

The EL results of two polycrystalline PV panels after 12 years of operation.

Module No. 1		Module No. 2
EL image	Specification	EL image	Specification
	$I_{s c}$ ⁠: 3.2477 A $I_{m p}$ ⁠: 2.9572 A $V_{m p}$ ⁠: 17.5093 V $P_{m a x}$ ⁠: 51.7783 W $V_{o c}$ ⁠: 22.785 V FF: 0.699		$I_{s c}$ ⁠: 3.0633 A $I_{m p}$ ⁠: 2.8703 A $V_{m p}$ ⁠: 17.5056 V $P_{m a x}$ ⁠: 50.2469 W $V_{o c}$ ⁠: 22.3169 V FF: 0.735%

Module No. 1		Module No. 2
EL image	Specification	EL image	Specification
	$I_{s c}$ ⁠: 3.2477 A $I_{m p}$ ⁠: 2.9572 A $V_{m p}$ ⁠: 17.5093 V $P_{m a x}$ ⁠: 51.7783 W $V_{o c}$ ⁠: 22.785 V FF: 0.699		$I_{s c}$ ⁠: 3.0633 A $I_{m p}$ ⁠: 2.8703 A $V_{m p}$ ⁠: 17.5056 V $P_{m a x}$ ⁠: 50.2469 W $V_{o c}$ ⁠: 22.3169 V FF: 0.735%

The calculation results of the degradation rate of polycrystalline PV panels are presented in Table 7. The data indicate that, after 12 years of operation, the degradation rate ranges from 1.32% to 1.62%. Additionally, to analyze the causes of this degradation, the EL imaging results of two polycrystalline PV panels after 12 years of operation reveal that the primary degradation is due to disconnections between cells.

Table 7.

Degradation of polycrystalline PV panels after 12 years of operation.

Panel NO.	Power (W)	Degradation power (W)	Degradation (%)	Average degradation (%)
No.1	51.778	8.222	15.88	1.32
No.2	50.246	9.754	19.41	1.62

Panel NO.	Power (W)	Degradation power (W)	Degradation (%)	Average degradation (%)
No.1	51.778	8.222	15.88	1.32
No.2	50.246	9.754	19.41	1.62

Table 7.

Degradation of polycrystalline PV panels after 12 years of operation.

Panel NO.	Power (W)	Degradation power (W)	Degradation (%)	Average degradation (%)
No.1	51.778	8.222	15.88	1.32
No.2	50.246	9.754	19.41	1.62

Panel NO.	Power (W)	Degradation power (W)	Degradation (%)	Average degradation (%)
No.1	51.778	8.222	15.88	1.32
No.2	50.246	9.754	19.41	1.62

The EL analysis results, as illustrated in Fig. 4, highlight the presence of oxidation at connection points, which leads to an increase in series resistance (R_s). Additionally, the accumulation of contaminants such as dust and animal waste (e.g. bird droppings) firmly adhering to the glass surface further exacerbates performance losses. Prolonged exposure to high ambient temperatures and humidity accelerates these degradation mechanisms by promoting water vapor infiltration beneath the glass surface. This infiltration fosters further oxidation and eventual disconnection of the metal strips that electrically connect the PV cells, resulting in an increase in R_s at the affected points.

Figure 4.

Analysis results using EL imaging of polycrystalline PV panels.

These environmental and structural factors collectively diminish the overall efficiency of the system. Performance indicators such as FF, V_oc, and degradation rate are directly impacted by these defects. Table 6 and Fig. 4 demonstrates how oxidation in polycrystalline modules leads to a marked increase in R_s, which, in turn, reduces both V_oc and FF. This interplay between environmental stressors, defect formation, and performance indicators underscores the critical need for robust maintenance strategies and defect monitoring to mitigate efficiency losses in PV modules.

3.3 Performance degradation analysis of monocrystalline PV panel

The measurement results obtained using EL technology for two monocrystalline PV panels with a rated power output of 110 Wp after 5 years of operation are presented in Table 8. According to Table 8, after 5 years of operation, the remaining power output is 100.245 W/110 W and 102.243 W/110 W, respectively. The corresponding FFs are 0.7598 and 0.7688. Table 9 presents the calculated degradation rates of the monocrystalline PV panels over the 5-year period. The results indicate that the annual degradation rate ranges from 0.282% to 0.354%, with an overall average degradation rate of 0.861% to 0.886% per year.

Table 8.

The EL results of two monocrystalline PV panels after 5 years of operation.

Module No. 1		Module No. 2
EL Image	Specification	EL Image	Specification
	$I_{s c}$ ⁠: 5.58 A $I_{m p}$ ⁠: 5.102 A $V_{m p}$ ⁠: 19.662 V $P_{m a x}$ ⁠: 100.245 W $V_{o c}$ ⁠: 22.882 V FF: 0.7598%		$I_{s c}$ ⁠: 6.045 A $I_{m p}$ ⁠: 6.0024 A $V_{m p} : 17.894$ V $P_{m a x}$ ⁠: 102.423 W $V_{o c}$ ⁠: 20.524 V FF: 0.7688%

Module No. 1		Module No. 2
EL Image	Specification	EL Image	Specification
	$I_{s c}$ ⁠: 5.58 A $I_{m p}$ ⁠: 5.102 A $V_{m p}$ ⁠: 19.662 V $P_{m a x}$ ⁠: 100.245 W $V_{o c}$ ⁠: 22.882 V FF: 0.7598%		$I_{s c}$ ⁠: 6.045 A $I_{m p}$ ⁠: 6.0024 A $V_{m p} : 17.894$ V $P_{m a x}$ ⁠: 102.423 W $V_{o c}$ ⁠: 20.524 V FF: 0.7688%

Table 8.

The EL results of two monocrystalline PV panels after 5 years of operation.

Module No. 1		Module No. 2
EL Image	Specification	EL Image	Specification
	$I_{s c}$ ⁠: 5.58 A $I_{m p}$ ⁠: 5.102 A $V_{m p}$ ⁠: 19.662 V $P_{m a x}$ ⁠: 100.245 W $V_{o c}$ ⁠: 22.882 V FF: 0.7598%		$I_{s c}$ ⁠: 6.045 A $I_{m p}$ ⁠: 6.0024 A $V_{m p} : 17.894$ V $P_{m a x}$ ⁠: 102.423 W $V_{o c}$ ⁠: 20.524 V FF: 0.7688%

Module No. 1		Module No. 2
EL Image	Specification	EL Image	Specification
	$I_{s c}$ ⁠: 5.58 A $I_{m p}$ ⁠: 5.102 A $V_{m p}$ ⁠: 19.662 V $P_{m a x}$ ⁠: 100.245 W $V_{o c}$ ⁠: 22.882 V FF: 0.7598%		$I_{s c}$ ⁠: 6.045 A $I_{m p}$ ⁠: 6.0024 A $V_{m p} : 17.894$ V $P_{m a x}$ ⁠: 102.423 W $V_{o c}$ ⁠: 20.524 V FF: 0.7688%

Table 9.

Degradation of monocrystalline PV panels after 5 years of operation.

Panel no.	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	110	100.245	9.755	1.951	0.354	0.861
No.2	110	102.243	7.757	1.551	0.282	0.886

Panel no.	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	110	100.245	9.755	1.951	0.354	0.861
No.2	110	102.243	7.757	1.551	0.282	0.886

Table 9.

Degradation of monocrystalline PV panels after 5 years of operation.

Panel no.	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	110	100.245	9.755	1.951	0.354	0.861
No.2	110	102.243	7.757	1.551	0.282	0.886

Panel no.	$P_{r e f}$ (W)	$P_{E L}$ (W)	Degradation power (W)	Degradation (%)	Average degradation (%)	Fill factor (%)
No.1	110	100.245	9.755	1.951	0.354	0.861
No.2	110	102.243	7.757	1.551	0.282	0.886

The EL images of the monocrystalline solar panel, as shown in Fig. 5, reveal performance degradation caused by defects such as micro-cracks and folds, which create shaded areas and reduce the panel’s ability to convert solar energy into electricity. Overlapping busbars and swollen solder joints further obstruct the current flow, limiting energy transmission. These defects not only arise from manufacturing issues but are also significantly influenced by environmental factors:

Figure 5.

Analysis results using EL imaging of polycrystalline PV panels.

Thermal expansion: When the solar panel operates under prolonged sunlight exposure, the metal components expand and contract due to thermal cycling. This phenomenon causes deformation of the busbars and solder joints, leading to overlaps and swelling, compromising the stability of electrical connections.

Oxidation and corrosion: In high-humidity environments, solder joints are prone to oxidation or corrosion, weakening the solder material and causing swelling. This increases resistance at connection points, negatively affecting current flow.

Additionally, the accumulation of dirt on the surface casts shadows over the PV cells, reducing light absorption efficiency and hindering the energy conversion process. These combined factors not only degrade the system’s performance but also shorten the lifespan of the solar panel, impacting the overall operational efficiency.

3.4 Degradation rates and statistical significance of different PV technologies

The statistical analysis results, presented in Table 10, reveal significant differences in degradation rates among the three PV technologies, as determined by analysis of variance (F(2, X) = Y, P < .05). These findings confirm the variability in degradation patterns and highlight distinct levels of resilience across the technologies under similar environmental conditions. Tukey’s post-hoc analysis further indicated that thin-film panels (mean = 1.45%, standard deviation = 0.05%) exhibited significantly higher degradation rates compared to monocrystalline panels (mean = 0.87%, standard deviation = 0.02%), with P < .01. Likewise, the degradation rate of polycrystalline panels (mean = 1.47%, standard deviation = .07%) was significantly higher than that of monocrystalline panels (P < .01). However, no statistically significant difference was found between the degradation rates of thin-film and polycrystalline panels (P = .12). These results underscore the superior resistance of monocrystalline panels to degradation relative to thin-film and polycrystalline panels under comparable environmental conditions.

Table 10.

Degradation rates and statistical significance of different PV technologies.

Technology	Mean degradation rate (%)	Standard deviation (%)	Significance (P-value)
Thin-film	1.45	0.05
Monocrystalline	0.87	0.02	P < .01 (vs. thin-film)
Polycrystalline	1.47	0.07	P = .12 (vs. thin-film)

Technology	Mean degradation rate (%)	Standard deviation (%)	Significance (P-value)
Thin-film	1.45	0.05
Monocrystalline	0.87	0.02	P < .01 (vs. thin-film)
Polycrystalline	1.47	0.07	P = .12 (vs. thin-film)

Table 10.

Degradation rates and statistical significance of different PV technologies.

Technology	Mean degradation rate (%)	Standard deviation (%)	Significance (P-value)
Thin-film	1.45	0.05
Monocrystalline	0.87	0.02	P < .01 (vs. thin-film)
Polycrystalline	1.47	0.07	P = .12 (vs. thin-film)

Technology	Mean degradation rate (%)	Standard deviation (%)	Significance (P-value)
Thin-film	1.45	0.05
Monocrystalline	0.87	0.02	P < .01 (vs. thin-film)
Polycrystalline	1.47	0.07	P = .12 (vs. thin-film)

3.5 Analysis of common errors during the production stage

In the PV module manufacturing process, various defects can arise, impacting module performance. These defects can be categorized into those visible to the naked eye and those requiring advanced detection methods like EL imaging. Visually observable defects include interlayer absence, compromised edge seals, uncured adhesive, loose frame screws, misaligned center bars, and white spots. EL imaging can reveal less apparent issues such as shading, small shunts, dark cell, and layer creases. These defects significantly affect module performance in different ways. Shunt damage creates localized areas of reduced current flow, limiting overall energy generation efficiency. Layer creases can cause micro-shading, directly impacting the cells’ photon absorption and energy conversion capabilities. Temperature-induced shading, considered a high-risk factor for performance degradation, can progressively expand, creating dark regions that encompass multiple cells and significantly reduce the module’s power output. These issues, particularly those related to shading and current flow disruption, can lead to non-uniform performance across the module, potentially resulting in hotspots and accelerated degradation over time. Table 11 provides an analysis of common defects during the manufacturing stage using EL imaging.

Table 11.

Common defects during the manufacturing stage using EL imaging.

Type of photovoltaic technology	Error detected	EL images
Thin-film	Shading: The shading size is approximately one-quarter, with the shadow equal to ≈2% of the cell length. The module is damaged when short-circuited, resulting in reduced efficiency.
	Shunt: Dark cells form small shunts, causing reduced current and power losses.
	Layer creases: Multiple-layer creases in PV panels cause significant losses by obstructing light, creating micro-shading, and disrupting current flow. They lead to hotspots, material degradation, and moisture ingress. Over time, these issues compound, substantially reducing overall module efficiency.
Polycrystalline	Disconnection: Over time, under the influence of high temperature and humidity, oxidation occurs, leading to the corrosion of metal strips. The formation of R_s resistance at these connection points impairs electrical conductivity, resulting in disrupted connections between the cells. This process not only affects the system’s performance but can also cause severe damage if not addressed in a timely manner.
Monocrystalline	Overlapping busbars and swollen solder: Prolonged sunlight exposure causes metal components in solar panels to expand and contract, deforming busbars, and solder joints, leading to overlaps and swelling that weaken connections. In high-humidity environments, oxidation and corrosion further degrade solder joints, increasing resistance at connection points and hindering current flow, ultimately reducing system performance.

Type of photovoltaic technology	Error detected	EL images
Thin-film	Shading: The shading size is approximately one-quarter, with the shadow equal to ≈2% of the cell length. The module is damaged when short-circuited, resulting in reduced efficiency.
	Shunt: Dark cells form small shunts, causing reduced current and power losses.
	Layer creases: Multiple-layer creases in PV panels cause significant losses by obstructing light, creating micro-shading, and disrupting current flow. They lead to hotspots, material degradation, and moisture ingress. Over time, these issues compound, substantially reducing overall module efficiency.
Polycrystalline	Disconnection: Over time, under the influence of high temperature and humidity, oxidation occurs, leading to the corrosion of metal strips. The formation of R_s resistance at these connection points impairs electrical conductivity, resulting in disrupted connections between the cells. This process not only affects the system’s performance but can also cause severe damage if not addressed in a timely manner.
Monocrystalline	Overlapping busbars and swollen solder: Prolonged sunlight exposure causes metal components in solar panels to expand and contract, deforming busbars, and solder joints, leading to overlaps and swelling that weaken connections. In high-humidity environments, oxidation and corrosion further degrade solder joints, increasing resistance at connection points and hindering current flow, ultimately reducing system performance.

Table 11.

10.1016/0960-1481(92)90064-a

Common defects during the manufacturing stage using EL imaging.

Type of photovoltaic technology	Error detected	EL images
Thin-film	Shading: The shading size is approximately one-quarter, with the shadow equal to ≈2% of the cell length. The module is damaged when short-circuited, resulting in reduced efficiency.
	Shunt: Dark cells form small shunts, causing reduced current and power losses.
	Layer creases: Multiple-layer creases in PV panels cause significant losses by obstructing light, creating micro-shading, and disrupting current flow. They lead to hotspots, material degradation, and moisture ingress. Over time, these issues compound, substantially reducing overall module efficiency.
Polycrystalline	Disconnection: Over time, under the influence of high temperature and humidity, oxidation occurs, leading to the corrosion of metal strips. The formation of R_s resistance at these connection points impairs electrical conductivity, resulting in disrupted connections between the cells. This process not only affects the system’s performance but can also cause severe damage if not addressed in a timely manner.
Monocrystalline	Overlapping busbars and swollen solder: Prolonged sunlight exposure causes metal components in solar panels to expand and contract, deforming busbars, and solder joints, leading to overlaps and swelling that weaken connections. In high-humidity environments, oxidation and corrosion further degrade solder joints, increasing resistance at connection points and hindering current flow, ultimately reducing system performance.

Type of photovoltaic technology	Error detected	EL images
Thin-film	Shading: The shading size is approximately one-quarter, with the shadow equal to ≈2% of the cell length. The module is damaged when short-circuited, resulting in reduced efficiency.
	Shunt: Dark cells form small shunts, causing reduced current and power losses.
	Layer creases: Multiple-layer creases in PV panels cause significant losses by obstructing light, creating micro-shading, and disrupting current flow. They lead to hotspots, material degradation, and moisture ingress. Over time, these issues compound, substantially reducing overall module efficiency.
Polycrystalline	Disconnection: Over time, under the influence of high temperature and humidity, oxidation occurs, leading to the corrosion of metal strips. The formation of R_s resistance at these connection points impairs electrical conductivity, resulting in disrupted connections between the cells. This process not only affects the system’s performance but can also cause severe damage if not addressed in a timely manner.
Monocrystalline	Overlapping busbars and swollen solder: Prolonged sunlight exposure causes metal components in solar panels to expand and contract, deforming busbars, and solder joints, leading to overlaps and swelling that weaken connections. In high-humidity environments, oxidation and corrosion further degrade solder joints, increasing resistance at connection points and hindering current flow, ultimately reducing system performance.

4. Experimental results analysis

4.1 Defect analysis based on experimental results

The experimental results obtained from EL imaging provide detailed insights into the nature and impact of defects across different PV technologies:

Thin-film modules: The analysis revealed shading, shunts, and layer creases as the primary defects. Shading was observed in areas with reduced EL intensity, indicating obstructed current flow. Shunts, identified as small localized dark regions in the EL images, significantly contributed to power losses by creating alternate current paths. Additionally, layer creases, visible as elongated dark streaks, obstructed light absorption and caused uneven current distribution, leading to hotspots and accelerated degradation. These defects collectively resulted in an average degradation rate of 7%–8% over 5 years, with some panels exhibiting as high as 18.5% degradation due to structural weaknesses or exposure to extreme environmental conditions.

Polycrystalline modules: The primary defects included oxidation-induced disconnections at interconnection points. The EL images showed regions of diminished intensity corresponding to areas with increased series resistance (R_s). The oxidation of metal strips, exacerbated by high humidity and temperature, led to disrupted electrical conductivity and reduced efficiency. Over 12 years, these defects resulted in degradation rates ranging from 1.32% to 1.62%. The presence of debris such as dust and animal waste further contributed to the performance losses by blocking sunlight and increasing the rate of material degradation.

Monocrystalline modules: Defects such as overlapping busbars and swollen solder joints were identified as key issues. EL images displayed dark regions along the busbars, corresponding to areas of increased resistance caused by thermal expansion and mechanical stress. In high-humidity environments, oxidation further weakens the solder joints, increasing resistance at connection points. Despite these challenges, monocrystalline modules showed the lowest degradation rates among the three technologies, ranging from 0.861% to 0.886% annually, highlighting their superior resilience under similar environmental conditions.

4.2 Correlation between defects and degradation rates

The quantitative analysis of the experimental results establishes a strong correlation between identified defects and degradation rates:

Thin-film panels with significant shading and shunt-related losses showed the highest degradation rates, particularly in panels with extensive layer creases.

Polycrystalline panels demonstrated moderate degradation rates, primarily influenced by oxidation and material contamination at interconnection points.

Monocrystalline panels exhibited minimal degradation due to the relatively lower prevalence and severity of defects, supported by their robust structural properties.

The statistical analysis (ANOVA) further confirmed significant differences in degradation rates among the technologies (P < .05). Monocrystalline modules consistently outperformed thin-film and polycrystalline technologies in resisting performance losses attributed to defects.

4.3 Mechanisms of defect-induced degradation

The mechanisms by which defects impact PV performance, as evidenced by the experimental results, are as follows:

Shading and shunts in thin-film modules: Reduce current flow, causing uneven energy distribution and localized heating, which accelerates material degradation.

Oxidation in polycrystalline modules: Increases Rs at interconnection points, disrupting electrical flow and creating inefficiencies that significantly reduce power output.

Overlapping busbars and swollen solder joints in monocrystalline modules: Increase resistance at connection points, weakening electrical conductivity and reducing overall efficiency.

5. Discussion

The analysis of performance degradation across different PV technologies under tropical monsoon climate conditions in Vietnam reveals that thin-film technology experiences an average degradation rate of 1.39%–1.53% over 5 years. This degradation is primarily attributed to the formation of small shunts, caused by high temperatures and humidity, which obstruct current flow and reduce power output. In more severe cases, these shunts enlarge, creating significant shading that further reduces efficiency. Notably, in one panel, the degradation reached as high as 18.526%, underscoring the profound impact environmental factors can have. These findings align with Chala et al. [26], who emphasized that temperature and humidity are critical factors affecting the performance and lifespan of PV modules.

Similarly, Wang et al. [27] demonstrated that temperature and humidity exacerbate PV module degradation, particularly in systems exposed to prolonged high-temperature and high-humidity cycles. These conditions compromise the durability of PV glass, leading to reduced light transmittance and overall structural integrity. Thin-film modules are especially vulnerable to environmental degradation compared to crystalline silicon technologies, exhibiting higher power loss rates over time when exposed to fluctuating temperature and humidity levels.

Polycrystalline technology shows an annual degradation rate ranging from 1.32% to 1.62% over 12 years, while monocrystalline panels have a lower degradation rate, ranging from 0.861% to 0.886% over 5 years. For polycrystalline modules, the main issues are cell disconnections due to oxidation and the accumulation of dust and animal waste on panel surfaces, which significantly reduce electrical connections and overall efficiency. The degradation observed in polycrystalline panels highlights the importance of regular maintenance and cleaning to extend their operational lifespan.

Statistical analysis confirmed significant differences in degradation rates between the studied technologies. Monocrystalline panels exhibited the lowest degradation rates, significantly lower than both thin-film and polycrystalline panels. This suggests that monocrystalline technology may offer superior long-term performance under similar environmental conditions. The absence of a significant difference between thin-film and polycrystalline panels suggests that both technologies share similar degradation mechanisms, likely due to their structural composition and environmental exposure. These findings point to the need for further research into mitigating degradation in both polycrystalline and thin-film technologies, especially in environments with high humidity and temperature fluctuations.

Furthermore, EL imaging has proven effective in detecting defects across different PV technologies, particularly in identifying shunts that cause micro-shading and obstruct current flow. These results are consistent with the findings of Puranik et al. [16], Deitsch et al. [11], and Koester et al. [22], all of which highlight the accuracy and low error rates associated with EL technology in detecting defects that affect module performance.

6. Conclusion

This study demonstrates that the quantitative EL method is a valuable tool for accurately assessing performance and detecting hidden defects in PV modules. Furthermore, it provides a reliable basis for optimizing maintenance strategies, especially in challenging tropical environments. The analysis results show that monocrystalline panels have the lowest degradation rate compared to polycrystalline and thin-film technologies, ranging from 0.861% to 0.886%. The main defects of thin-film technology are the formation of small shunts that gradually accumulate and form shading, obstructing current flow and reducing efficiency. For monocrystalline and polycrystalline technologies, the defects include oxidation leading to loss of connection, layer wrinkles creating shading, as well as dust accumulation and animal waste. The study also demonstrates the effectiveness of using EL to identify micro-defects with high accuracy.

Author contributions

Le Truong (Conceptualization [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Nguyen Liet (Data curation [equal], Resources [equal], Writing—review & editing [equal]).

Conflict of interest

None declared.

Data Availability

Data can be available on reasonable request.

References

[1]

Som

Al-Alawi

SM.

Evaluation of efficiency and degradation of mono- and poly-crystalline PV modules under outdoor conditions

Renew Energy

1992

;

–

. https://doi.org/

10.1016/j.physb.2009.09.011

[2]

Radue

van Dyk

Degradation analysis of thin film photovoltaic modules

Physica B

2009

;

404

4449

–

. https://doi.org/

10.1016/j.solener.2016.10.030

[3]

Silvestre

Kichou

Guglielminotti

et al.

Degradation analysis of thin film photovoltaic modules under outdoor long-term exposure in Spanish continental climate conditions

Sol Energy

2016

;

139

599

–

607

. https://doi.org/

[4]

Smith

Jordan

Kurtz

Outdoor PV Module Degradation of Current-Voltage Parameters: Preprint

Golden

National Renewable Energy Lab (NREL)

2012

Google Preview

OpenURL Placeholder Text

10.1016/j.solener.2016.06.047

[5]

Rajput

Venkateswaran

Bora

et al.

Degradation of mono-crystalline photovoltaic modules after 22 years of outdoor exposure in the composite climate of India

Sol Energy

2016

;

135

786

–

. https://doi.org/

10.1109/IESC.2018.8439990

[6]

Faye

Philippe

Ndiaye

et al.

Degradation and comparative experimental study of crystalline photovoltaic module after a few years outdoor exposure in Casamance and Cologne climate

. In: 2018 7th International Energy and Sustainability Conference (IESC), Cologne,

May 2018

, pp.

–

New York

IEEE

. https://doi.org/

[7]

Singh

Sharma

Yadav

Degradation and reliability analysis of photovoltaic modules after operating for 12 years: a case study with comparisons

Renew Energy

2022

;

196

1170

–

. https://doi.org/

10.1016/j.renene.2022.07.048

10.1016/j.solmat.2024.113118

[8]

Abdallah

Abdelrahim

Elgaili

et al.

Degradation of photovoltaic module backsheet materials in desert climate

Sol Energy Mater Sol Cells

2024

;

277

113118

. https://doi.org/

[9]

Lai

Potter

Simmons-Potter

EL image analysis of a photovoltaic module under accelerated lifecycle testing

Appl. Optics

2020

;

5656

–

. https://doi.org/

[10]

Njok

Ogbulezie

The effect of relative humidity and temperature on polycrystalline solar panels installed close to a river

Phys Sci Int J

2019

;

–

. https://doi.org/

10.9734/PSIJ/2018/44760

10.1016/j.solener.2019.02.067

[11]

Deitsch

Christlein

Berger

et al.

Automatic classification of defective photovoltaic module cells in electroluminescence images

Sol Energy

2019

;

185

455

–

. https://doi.org/

10.1109/PVSC40753.2019.8981398

[12]

Seigneur

Schneller

Morlier

et al.

Electroluminescence-testing induced crack closure in PV modules

. In: 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC), Chicago, IL, pp.

863

–

2019

New York

IEEE

. https://doi.org/

[13]

Lai

Ramirez

Potter

Jr et al.

Image analysis of PV module electroluminescence

. In:

Reliability of Photovoltaic Cells, Modules, Components, and Systems X

, ,Vol.

10370

Washington, DC

SPIE

2017

129

–

. https://doi.org/

10.1117/12.2274184

Google Preview

OpenURL Placeholder Text

10.1109/JPHOTOV.2018.2846665

[14]

Bedrich

Luo

Pravettoni

et al.

Quantitative electroluminescence imaging analysis for performance estimation of PID-influenced PV modules

IEEE J Photovoltaics

2018

;

1281

–

. https://doi.org/

10.1109/IPCon.2017.8116193

[15]

Bazzoli

Silverman

Goddard

LL.

Low-cost electroluminescence imaging for automated defect characterization in photovoltaic

modules

. In: 2017 IEEE Photonics Conference (IPC), Orlando, FL,

October 2017

, pp.

481

–

New York

IEEE

. https://doi.org/

[16]

Puranik

Kumar

Gupta

Generalized quantitative electroluminescence method for the performance evaluation of defective and unevenly degraded crystalline silicon photovoltaic modules

Prog Photovolt Res Appl

2023

;

269

–

. https://doi.org/

[17]

Liu

Wang

et al.

An efficient CNN-based detector for photovoltaic module cells defect detection in EL images

Sol Energy

2024

;

267

112245

. https://doi.org/

10.1016/j.solener.2023.112245

10.1016/j.eswa.2024.123199

[18]

Liu

Jonathan

et al.

A real-time anchor-free defect detector with global and local feature enhancement for surface defect detection

Expert Syst Appl

2024

;

246

123199

. https://doi.org/

10.1016/j.rser.2014.01.027

[19]

Ciulla

Brano

Di Dio

et al.

A comparison of different one-diode models for the representation of I–V characteristic of a PV cell

Renew Sustain Energy Rev

2014

;

684

–

. https://doi.org/

10.1016/j.solener.2023.111994

[20]

Wang

Sun

et al.

Deep-learning-based automatic detection of photovoltaic cell defects in EL images

Sensors

2023

;

297

. https://doi.org/

[21]

Puranik

Kumar

Gupta

Progress in module-level quantitative EL imaging of crystalline silicon PV modules: a review

Sol Energy

2023

;

264

111994

. https://doi.org/

10.1016/j.rser.2022.112616

[22]

Koester

Lindig

Louwen

et al.

Review of photovoltaic module degradation, field inspection techniques and techno-economic assessment

Renew Sustain Energy Rev

2022

;

165

112616

. https://doi.org/

10.1002/j.1538-7305.1949.tb03645.x

[23]

Shockley

The theory of p-n junctions in semiconductors and p-n junction transistors

Bell Syst Tech J

1949

;

435

–

. https://doi.org/

10.1109/PVSC.2005.1488387

[24]

Meyer

Van Dyk

EE.

The effect of reduced shunt resistance and shading on photovoltaic module performance

. In: Conference Record of the Thirty-first IEEE Photovoltaic Specialists Conference, Lake Buena Vista, FL, pp.

1331

–

2005

New York

IEEE

. https://doi.org/

[25]

First Solar Inc

Series 6 Shading Risk Assessment

Ho Chi Minh City, Vietnam

2024

Google Preview

OpenURL Placeholder Text

10.1016/j.heliyon.2024.e30944

[26]

Chala

Sulaiman

Al Alshaikh

SM.

Effects of climatic conditions of Al Seeb in Oman on the performance of solar photovoltaic panels

Heliyon

2024

;

e30944

. https://doi.org/

[27]

Wang

Bao

Qiu

et al.

Durability of photovoltaic laminated glass modules affected by temperature-humidity cycles

Key Eng Mater

2016

;

680

323

–

. https://doi.org/

10.4028/www.scientific.net/kem.680.323