Review of I–V Electrical Characterization Techniques for Photovoltaic Modules Under Real Installation Conditions

Abstract

The exploitation and development of photovoltaic (PV) modules faces several technical challenges, including those related to variability in electrical performance under real conditions, such as temperature fluctuations, irradiance variability, and dust accumulation. One solution for evaluating and controlling these performances is to conduct electrical characterization under natural conditions. Many characterization techniques have been developed and proposed in the literature, with the aim of verifying manufacturer performance guarantees and better understanding the behavior of PV modules in their installation environment, where the climatic parameters, such as solar irradiation and temperature, fluctuate constantly. These techniques are based on recognized standards, including those established by the International Electrotechnical Commission (IEC) and American Society for Testing and Materials (ASTM). They are also based on methods of transposing basic electrical parameters, allowing the prediction of the performance of modules under various environmental conditions. In this work, a classification and a critical analysis of the main methods of electrical characterization were undertaken, highlighting their respective advantages and disadvantages. The experimental protocols used to evaluate the impact of environmental parameters on the performance of PV modules were examined in detail.

1. Introduction

Photovoltaic (PV) technologies are becoming increasingly important for energy production [1,2,3,4,5,6,7]. However, the precise prediction of their performance under real conditions remains a major issue, especially since most manufacturers offer warranties of 25 to 30 years with a power decline of less than 20% during this period [8,9]. The reliability and longevity of PV modules are mainly determined by their energy performance [10,11], which is strongly influenced by environmental factors such as temperature and solar irradiation. Other factors, such as dust accumulation and degradation, also play important roles. Dust fouling of the modules reduces the received irradiance and increases their operating temperature, which can accelerate degradation mechanisms.

In this context, fast and accurate measurement methods are imperative for evaluating the performance of photovoltaic generators. These not only quantify the influence of internal and external parameters on electrical performance but also adapt PV systems to real operating environments [12,13]. Indeed, PV modules rarely operate under the standard test conditions (STC) specified by manufacturers. Therefore, it is necessary to collect reliable data, such as I–V curves, output power, and degradation rate, on their behavior under real conditions, in order to improve their operation and design.

One of the most widely used techniques for this purpose is current–voltage (I–V) characterization, which allows for understanding the electrical behavior of PV modules under indoor and outdoor conditions. It provides essential information for improving the reliability of photovoltaic power generation and evaluating the efficiency of different PV technologies [14,15].

Characterization methods can be classified into two main categories: indoor and outdoor. Indoor characterization is carried out in the laboratory using solar simulators (often of the flasher type) and climatic chambers, allowing precise temperature control [16]. Although this method offers high reproducibility, it requires expensive equipment, such as solar simulators, and maintaining indoor air conditioning does not reproduce the results with the same fidelity as in real environmental conditions [17]. Outdoor characterization, carried out under natural conditions, allows the evaluation of the performance of PV modules in their real operating environment. However, the variability of weather conditions and availability of data represent significant challenges that some researchers are attempting to overcome through numerical simulations [18].

The diversity of characterization approaches depends on the specific objectives of each study, as well as the climatic and electrical variables taken into account. For example, the efficiency of PV modules decreases differently depending on the technology, especially when the irradiance drops below 200 W/m², a scenario rarely considered under STC conditions, where irradiation is set at 1000 W/m² [19]. To better understand this phenomenon, Mambrini, et al. [20] implemented an experimental method for correcting the spectral shift between real solar radiation and the reference spectrum AM1.5 (ASTMG), based on several weeks of outdoor measurements.

Recent work has also focused on the impact of dust on PV module performance. For example, Dhaouadi, et al. [21] developed a characterization method that takes into account the nature and physical properties of dust collected in Sharjah (UAE), while Sadat, et al. [22] used scanning electron microscopy, X-fluorescence, and elemental mapping to analyze the chemical composition and morphology of dust particles. They found that these particles, ranging in size from 0.4 μm to 31 μm, mainly contained silicon, oxygen, aluminum, and calcium. (Kazem et al. [23] proposed an experimental model to estimate the impact of dust on grid-connected photovoltaic systems.

In addition, Gadong, et al. [24] relied on measurements of fundamental electrical parameters (I_sc, V_oc, P_max) to characterize crystalline PV modules using a system with variable resistive load. Other authors, such as Silva, et al. [14], have proposed a model that takes into account UV radiation, visible irradiation, ambient temperature, and module temperature. Their method, based on the capacitive charge technique, uses a device connected to Wi-Fi for the real-time transmission of data to a computer.

This work offers a critical review of the different techniques for the electrical characterization of photovoltaic modules used under real operating conditions. The aim is to identify the approaches developed in the literature, to analyze their methods, experimental devices, and protocols used, as well as to assess their relevance with regard to the study of the performance of PV modules in relation to real environmental conditions.

Although several publications have focused on I–V characterization techniques [25,26,27,28], to the best of our knowledge, no comprehensive review has focused on the detailed description of the experimental protocols and materials used in real conditions. This work, therefore, fills this gap by providing a complete panorama of outdoor characterization techniques.

Section 2 presents the different techniques of electrical characterization of PV modules, in laboratory (indoor) and outdoor environments, as well as the methods and protocols adopted by the authors to analyze the effect of environmental parameters. Section 3 describes the techniques for transposing measured parameters to standard test conditions (STC).

2. Characterization Techniques for PV Modules

2.1. Laboratory Characterization (Indoor)

2.1.1. Test Conditions and Data Sheets

The main characteristics of PV modules, determined in the laboratory under standard test conditions (STC), are often provided by manufacturers in data sheets. Some are obvious to understand, such as mass, dimensions, and electrical and mechanical characteristics. The electrical characteristics often encountered are the short circuit current (I_sc), open circuit voltage (V_oc), maximum current (Imp), maximum voltage (V_mp), maximum power (P_max), temperature coefficients, etc., recorded in standard test conditions (STC). Some data sheets also provide information on performance guarantees or on the limits of living conditions. In this work, we are interested in electrical data and characteristic curves as well as their evolution under real operating conditions. Manufacturers also often provide the nominal operating cell temperature (NOCT). This is the operating temperature of the cell or module at an illuminance of 800 W/m², ambient temperature of 20 °C, wind speed of 1 m/s, and spectrum AM1.5. NOCT allows, in particular, the calculation of the temperature of the cell (T_c) or PV module from the ambient temperature (T_a) and illumination (G) using the following formula:

$$T_c=T_a+\left({\displaystyle \frac{NOCT-20}{800}}\right)·G$$

(1)

2.1.2. The Main Types of Laboratory Characterization

In the context of the characterization of PV modules in the laboratory, solar simulators and other means of thermal control are used to adjust the test conditions according to specific needs. These procedures are essential for determining the characteristics of PV modules specified on nameplates and in data sheets, especially under standard test conditions (STC) and nominal operating temperature conditions (NOCT).

The solar simulator, usually in the form of a flash, is commonly used by manufacturers to measure the output performance of PV modules. It is a standard procedure to verify the correct functioning of each module to ensure compliance with the technical specifications. The control of test conditions, in particular, the temperature management of PV modules, is crucial for the precise determination of temperature coefficients, which are key parameters for evaluating module performance under varying conditions.

In addition, unlike natural sunlight, the spectral distribution in a solar simulator is more stable, resulting in better reproducibility of measurements. This makes laboratory testing more reliable, as noted by Minimoto, et al. [29]. Although several types of solar simulators exist, namely continuous, pulsed, or LED, they are classified according to specific criteria defined by the International Electrotechnical Commission (IEC) in standard IEC 60904-9 [30], as detailed in

Table 1. Classification of solar simulators according to standard 60904-9.

Quality Indicator	Method		Classification
		Class A	Class B	Class C
Non-uniformity of illumination	Monitoring the distribution of illumination in the test area. Calculation from measured min/max values of the illuminance	<2%	<5%	<10%
Match to the reference illuminance spectrum AM1.5 (IEC60904-3 [31])	Ratio of the illuminance inputs of six wavelength ranges (400, 500, 600, 700, 800, 900, 1100): solar simulator/reference AM1.5	0.75–1.5%	0.6–1.4%	0.4–2%
Temporal stability of the emitted light (LTI = Long-term Instability)	Tracking of illumination in a fixed position in the test area. Calculation from min/max values during the I–V data acquisition period.	<0.5%	<2%	<10%

.

Several approaches for the indoor characterization of PV modules have been explored in the literature in order to ensure optimal operating conditions [32,33]. Thus, Siddiqui, et al., 2014 [33] conducted laboratory tests on different monocrystalline and polycrystalline silicon PV modules. The objective of this study was to identify the simplest and most reliable method for obtaining current-voltage (I–V) curves of PV modules under STC conditions, taking into account the duration of exposure to sunlight. A calibrated solar simulator of the ENDEAS QUICKSUN 700A (Figure 1) type was used to determine the temperature and irradiance coefficients in order to ensure good reproducibility of the I–V curves in accordance with the IEC 61215 standard [34]. In order to stabilize the electrical and thermal parameters, the PV modules underwent preconditioning by exposing them to the sun for four hours. Measurements were then carried out immediately to maintain the activation state of the modules and ensure the reliability of the collected data. To ensure a module temperature close to 25 °C before the measurements, several water circulations were carried out using the refrigeration system. Once the temperature of the modules stabilized around 25 °C and the irradiance reached about 1000 W/m², four measurements were performed to verify the stability of the conditions. Although the conditions are very close to the STC, the measurements obtained were extrapolated to the STC using the irradiance and temperature coefficients determined previously.

Similarly, Pratt, et al. [35] conducted an inter-laboratory comparative study (Inter-Laboratory Comparison—ILC) evaluating the performance of crystalline silicon PV modules. The study involved three South African institutions and aimed to identify measurement gaps related to differences in experimental protocols, equipment, and calibration procedures. One of the major objectives was to stress the need for harmonization of characterization methods in order to ensure reliability and comparability of results between laboratories. The analysis made it possible to quantify the differences in power, current, and voltage measured between the different experimental configurations. This article also provides an overview of the fundamental principles of PV module performance measurements, their relationship to manufacturers’ specified nominal values, and a detailed description of the main sources of uncertainty associated with the measurements. In this study, tests were carried out using solar simulators with integrated thermal chambers. The results were then compared to those obtained from an Optosolar simulator equipped with a xenon lamp and an MBJ (Mary Beth and Jim Bos, a human resources talent management company) mobile tester using LED technology. Lalaoui, et al. [36] have carried out an in-depth performance assessment of PV modules under indoor conditions, using a solar simulator, with the aim of verifying the conformity of the measured electrical characteristics with those declared by the manufacturer, according to DIN EN ISO/IEC 17025:2005 [37]. Preliminary tests were conducted at the CIS SolarTestLab in Erfurt, Germany, using a solar simulator of the PSS 8-AU type manufactured by BERGER Lichttechnik (see Figure 2). During the tests, each PV module was exposed to a short-lived flash (3 ms) with an intensity of 100 mW/cm², emitted by a xenon arc lamp. The output signal was captured using a Pulsed Solar Load and Measuring Device (PSL SCD) measurement system, which simulates a computer-controlled variable resistance. The obtained data were compared with those of a reference solar module. The results showed a decrease in the measured performance from the manufacturer’s stated values: a decrease of 1.2% for mono-crystalline silicon (m–Si) modules, 3% for polycrystalline silicon (p–Si), and up to 10% for amorphous silicon (a–Si), indicating significant deviations depending on the technology used.

Zettl et al. [16] used an innovative solar simulator (ISS) to analyze the behavior of PV modules under various conditions. This device consists of three main elements: an irradiation source, an artificial cooling system (simulated cold), and a collecting surface used as a support for the mounting of the modules (see Figure 3). The tests focused on the influence of several key factors, including the angle of incidence, direct and diffuse irradiation, and the effect of temperature on the performance of PV modules.

The light spectrum emitted by the simulator lamps is designed to be close to the actual solar spectrum, which improves the measurement accuracy. The cooling system maintains the modules at a stable temperature close to ambient conditions. The PV modules are fixed on a frame mounted on the collector surface, which can be tilted between 0° and 90°, thus allowing the angle of incidence between the active surface of the module and the light source to be changed. The main objective of this study was to estimate the performance ratio (PR) of different types of PV modules, regardless of their orientation or location. Apostolou et al. [38] proposed a simplified comparative model to estimate the performance of 12 commercial PV modules under indoor conditions. This model makes it possible to predict the performance of PV modules based on their distance from different light sources, whether artificial (fluorescent, halogen, or LED) or natural. The approach adopted is based on the combined use of two light sources: natural light, which may be insufficient indoors, and appropriate artificial lighting. The main objective of this work is to provide a design aid tool for integrated photovoltaic systems, optimized for use in indoor environments. The model is based on physical measurements of both natural and artificial indoor irradiance. The main input data include: module surface, spectral response (SR) of the device, indoor spectral irradiance, and distance between solar cell and light sources. The analytical model, developed in Microsoft Excel, allows calculation of solar cell efficiency and power output under specific indoor conditions, as shown in Figure 4 [39].

Luciani et al. [40], performed an indoor characterization to determine the correction parameters of PV modules in accordance with IEC 60891 [41]. To do this, a set of nine (9) PV modules was tested using a flash test bench based on a class A+ solar simulator (AM 1.5), manufactured by BERGER Lichttechnik. The experimental device, shown in Figure 5, consists of a pulsed solar simulator (PSS), solar pulsed charge system (PSL), infrared detector (IR), temperature sensor Pt100, computer with dedicated acquisition software for recording current-voltage (I–V) curves, and tower positioning system. The solar simulator incorporates an energy generator and a lamellar light source without optical elements, thus ensuring homogeneous and reproducible lighting. The PSL system, with three channels controlled by a processor, allows for load simulation and precise measurement of PV module performance. This device conforms to IEC 60904-1 [42] for the measurement of I–V curves. The I–V curves obtained under standard test conditions (STC) after the application of the determined correction parameters were compared with those measured using a commercial curve plotter, as reported in an earlier study [2].

Ozcan, et al. [43] used a temperature-controlled solar simulator based on a xenon light source and equipped with radiation attenuator masks, in order to accurately analyze the effect of temperature on the electrical parameters of a single-crystal PV module. This simulator, shown in Figure 6, incorporates a closed thermal chamber with a quick-opening door designed to preserve the thermal uniformity on the surface of the tested modules. The spectral concordance of the simulator was evaluated using a set consisting of a spectroradiometer, optical incidence system, and dedicated software. To study the spatial distribution of the illumination provided by the simulator, an area of 22 m was divided into 64 equal segments, in accordance with the requirements of IEC 60904-9 [30]. The irradiance in each segment was measured using a specially designed PV cell (16 × 16 cm), positioned successively on each of the 64 zones, irradiated by the simulator, and then analyzed to extract the performance parameters. The electrical characteristics of a PV module made of monocrystalline silicon (m–Si) were measured at different illumination levels and temperatures. The temperature coefficients were evaluated using the interpolation method described in IEC 60891, based on a simple matrix linking irradiance and temperature.

Abella et al. [44] presented an experimental procedure for determining the temperature coefficients and correction parameters in accordance with the international standard IEC 60891. The coefficients α and β, respectively, are the temperature coefficients for the current and voltage, while ${\alpha }_{réel}$ and ${\beta }_{réel}$, the normalized temperature coefficients for the module current and voltage under real site conditions, as well as the parameters κ, κ′, R_s, R_s′ and a respectively curve correction factor which acts as a temperature coefficient for the series internal resistance R_s, curve correction factor which acts as a temperature coefficient for the series internal resistance R_s′, the series internal resistance R_s at STC, the series internal resistance R_s′ at real conditions and the curve correction factor interpreted as the thermal voltage of the diode, are determined from the current-voltage (I–V) curves measured in a solar simulator, at irradiance levels between 700 and 1100 W/m² and for temperatures ranging from 20 to 50 °C. The thermal regulation of the PV module is carried out in two stages. The first regulation is ensured by an air heating/cooling system, allowing the temperature of the black box containing the solar simulator to be stabilized between 20 °C and 30 °C. Then, to achieve module temperatures between 30 °C and 50 °C, forward biasing of the PV module is applied using a DC power supply operating at a voltage higher than the no-load voltage V_oc, while limiting the current to twice the rated value. Once the maximum temperature is reached, the power supply is disconnected, and the I–V curves are measured. The natural cooling of the module then allows data to be collected at different temperature levels. The results obtained for the temperature coefficients α, β, and at an irradiance of 1000 W/m² are shown in Figure 7.

The correction parameters k and k′, corresponding to correction procedures 1 and 2 of IEC 60891, respectively, as well as the series resistors R_s and R_s′, are calculated from the I–V curves measured at constant irradiance for different temperatures. These Calculations are based on the analysis of power deviations, as shown in Figure 8.

Dubey et al. [45] also proposed an experimental procedure for the determination of temperature coefficients based on characterization under indoor conditions using a solar simulator. To do this, the PV modules were first exposed to sunlight, with their output terminals short-circuited to accelerate their heating, due to Joule losses in the metallization and interconnections. Once heated, the modules were cleaned, water-cooled, wiped, and covered with opaque cardboard to allow for homogeneous stabilization of their temperature. When the surface temperature reached 60 °C (detected by an infrared camera, Building Diagnostics Group, New Orleans, LA, USA), the cardboard was removed, and the module was transferred inside the laboratory under a solar simulator (SPIRE SPI-SUN 5600 SLP BLUE, interpv.net Corporation, New Orleans, LA, USA), as shown in Figure 9. Using the laboratory’s air conditioners, the temperature of the PV module was then maintained at a constant value of 25 °C, corresponding to Standard Test Conditions. The simulator was programmed to record I–V curves at three different irradiance levels at 30 s intervals until the temperature of the module gradually decreased to 25 °C. The short-circuit currents (I_sc) and open-circuit voltages (V_oc) extracted from the I–V curves were plotted as a function of the module temperature, thus allowing the temperature coefficients to be estimated.

2.1.3. Discussion and Critical Analysis of Laboratory Characterization

Laboratory PV module characterization techniques are at the heart of technological validation, certification, and optimization processes for solar systems. They allow for the precise measurement of the electrical and thermal parameters of PV modules under controlled conditions. These techniques offer undeniable advantages.

One of the main strengths of laboratory tests is their reproducibility [29]. The measurements are carried out under standard test conditions (STC). These conditions allow for the comparison of different PPV modules on a common basis. The use of solar simulators, whether continuous, pulsed, xenon lamps, or LED, offers spectral stability much higher than that of natural light, which is subject to climatic variations. This equipment, classified according to the IEC 60904-9 standard, guarantees luminous uniformity and stability of illumination, which is essential for test accuracy.

From a technical point of view, laboratories have high-performance measuring instruments to establish I–V curves and evaluate the main electrical performance characteristics of PV modules. Ultra-fast flash tests (of the order of a few milliseconds) avoid the heating of PV cells during measurement, guaranteeing reliable results. In addition, active cooling systems ensure the thermal stability of PV modules during testing [16]. These devices are complemented by climatic chambers to test PV modules at different temperatures, which is essential for determining the temperature coefficients (α, β, and δ) that influence performance under real conditions [44].

Another notable advantage is the detection of gaps between announced and actual performance [38]. Some manufacturers overestimate the characteristics of their PV modules, and laboratory tests can highlight differences of up to 10%, particularly for certain technologies, such as amorphous silicon (a-Si). Inter-laboratory studies, such as that of Pratt et al. [35], contribute to harmonizing measurement practices and enhancing transparency in the photovoltaic industry.

However, while these techniques offer undeniable advantages, they also have practical, economic, and methodological limitations that merit in-depth analysis.

First, they are carried out under artificial conditions that do not accurately reflect the complexity of external environments. The light spectrum of simulators, even class A, remains imperfect, especially for technologies sensitive to certain wavelengths. In addition, the tests ignore real factors such as module fouling, partial shading, humidity, wind variations, and soil albedo, which strongly influence energy production under real conditions.

Another constraint is the high cost associated with sophisticated equipment. High-end solar simulators, climate chambers, precision sensors, and cooling systems represent a major investment, often inaccessible to developing countries. In addition, regular maintenance of these devices, including the replacement of lamps or calibration of sensors, is essential to maintain the reliability of measurements.

From a methodological point of view, some procedures require preconditioning of the modules (exposure of several hours) before the measurements, which extends the test times. The results obtained must then be extrapolated to standard test conditions via coefficients defined by IEC 60891, which introduces additional uncertainties. Finally, inter-laboratory variability persists despite the standards: human handling, orientation deviations, or cleaning of modules can affect the results.

Faced with these limitations, a hybrid approach combining laboratory tests and monitoring under real conditions seems relevant. Outdoor tests allow the observation of the behavior of modules in their operating environment, taking into account all natural variables. Simple monitoring systems (irradiance, temperature, and humidity sensors) allow long-term evaluation at a much lower cost. However, these measures suffer from a lack of control and reproducibility.

The prospects for improvement are based on several axes: the development of multi-spectral solar simulators capable of reproducing the natural solar spectrum with greater fidelity, the automation of tests to limit human biases, the creation of strengthened international protocols to improve the comparability of results, and the integration of intelligent algorithms to correct deviations related to ambient conditions.

In conclusion, laboratory characterization techniques play a fundamental role in the development and regulation of the photovoltaic industry. They ensure an accurate, standardized, and reproducible evaluation of the performance of the modules, but must be supplemented by tests in a real environment to reflect the true operating conditions. The future lies in the synergy between experimental rigor and field observation, supported by technological and normative innovations.

2.2. Characterization in Natural Conditions (Outdoor)

2.2.1. Current-Voltage (I–V) Characterization

Characterizing the photoelectric properties of a photovoltaic (PV) module consists of plotting its current–voltage (I–V) and power-voltage (P–V) curves under standard conditions (STC), in order to extract key electrical parameters such as short circuit current (I_sc), open circuit voltage (V_oc), maximum power (P_max), form factor (FF), and efficiency. Two approaches are used: laboratory measurement (indoor) and real-world measurement (outdoor).

The IEC 60904-1 standard, revised in 2020 [42], defines the general requirements for measuring I–V characteristics both indoors and outdoors. In the laboratory, the characterization is carried out using solar simulators and thermal control devices. At the site, various loads (resistive, capacitive, electronic, and inductive) or DC-DC converters are mobilized [25,26,46].

Recent studies have highlighted innovative approaches, notably the use of wireless and self-powered plotters [27], allowing for the rapid detection of panel failures and a detailed evaluation of their performance, including under partial shading [47]. Other works have delved into the effects of temperature and irradiance [48,49] or explored advanced techniques to improve measurement accuracy and responsiveness [50,51,52,53,54]. These advances make I–V characterization more suitable for performance monitoring under real conditions.

The different methods of I–V characterization under real conditions are summarized in Figure 10.

Resistive Load Characterization

The simplest technique for characterizing a PV module is to measure its current-voltage characteristics by connecting variable resistors to the output of the PV module [26,55,56]. The value of these resistors varies from zero (short circuit) to infinity (open circuit), thus allowing the entire I–V curve to be scanned. However, this method has limited accuracy because it is very sensitive to environmental variations during scanning, especially if performed manually. This manual scanning means that this technique suffers from a high variability. These factors significantly influence the test results.

To overcome this limitation, Amiry, et al. [57] proposed a technique for automatic variation of the load resistance. The experimental device used includes a variable resistance, an Arduino Mega 2560 board, and several suitable sensors: current, voltage, irradiance, and temperature sensors (see Figure 11). The electronic diagram of the assembly is shown in Figure 12 and Figure 13. The surface temperature of the module is measured using a K-type thermocouple, and the irradiance is recorded using a pyranometer positioned in the module plane.

To reduce the number of resistors and switches, Rivai, et al. [58] introduced a binary I–V curve plotter based on an adjustable resistance load. In this system, the resistors are connected in series, each in parallel with a switch (SWk). The resistance variation is ensured by the combination of openings and closures of these switches (SWn 1…SW0), which allows for obtaining an equivalent resistance adjustable by binary steps (see Figure 14). The performance of this method has been compared to that of the conventional method described by Dyk, et al. [2], with eight fixed resistances. The classical method generates eight points on the I–V curve, while the proposed tracer allows for obtaining up to 255 points. The latter showed comparable accuracy to the conventional method at 1000 W/m² irradiance, while maintaining this accuracy under low irradiance levels (below 600 W/m²) or in the case of partial shading of the PV generator, conditions in which the conventional method fails to identify the maximum power point (MPP). Willoughby et al. [55] used a sequential selection of relays, each connected to a fixed load resistance, to determine the different points of the I–V curve. The voltages, currents, and irradiance levels measured by a pyranometer are recorded simultaneously using a data logger. The main advantages of these tracers are their low cost, ease of implementation, and automation capability. Nevertheless, this technique has several limitations: sometimes insufficient curve quality, accuracy affected by fluctuations in solar radiation and temperature, and, in many cases, manual variation of resistances. In addition, it is better suited to low-power PV generators, as high-power resistors are rare and expensive.

Electronic Charge Characterization

This is a simple and inexpensive characterization technique [12,59,60,61]. Most authors exploit a Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET), which is a fundamental electronic component in digital electronics, particularly in the manufacture of microprocessors, where it is used for the construction of logic gates. This field-effect transistor modulates the drain-source current via a voltage applied to its grid. In the characterization of PV modules, the MOSFET is used as a variable electronic charge, whose gate voltage is controlled by an RC filter [60] or an RLC filter [61] in combination with an Arduino board (see Figure 15). The system originally proposed by Leite et al. [59] was subsequently improved by the same authors [12], integrating it with a low-cost data acquisition (DAQ) system, driven by an application developed under LabVIEW, which increased the flexibility and performance of the I–V plotter. The technique, based on an RLC circuit, allows the I–V and P–V curves to be obtained taking into account variations in external conditions such as temperature, irradiance, and dust, while facilitating the monitoring of the maximum power point (MPP) [61]. According to the authors, this method offers a faster response time (80 ms) than that using an RC filter (100 ms). The data is acquired by current and voltage sensors coupled to the Arduino board, and the data are then stored in real time in an Excel sheet via the Parallax PLX-DAQ macro (see Figure 16).

Other work, such as that of Kuai, et al. [62], implemented a test circuit based on a MOSFET operating in linear mode (see Figure 17), in order to continuously obtain the I–V and P–V curves to evaluate the performance of a photovoltaic installation. A block diagram of the device is shown in Figure 18.

However, this approach has its limitations. Indeed, due to the safe operating zone of the MOSFET in linear mode, where it behaves as a variable resistance, the component can only dissipate high power over very short durations (a few milliseconds). Therefore, MOSFET-based techniques are poorly suited for the characterization of high-power PV generators [63]. In addition, the high heat dissipation required to accurately reproduce the points around the MPP is a major disadvantage of electronic charge sensors, particularly when the transistor operates in its linear zone.

Capacitive Charge Characterization

The capacitive charge technique allows the electronic generation of the I–V and P–V characteristics, independently of the size of the photovoltaic field, and at a reduced cost [64]. It also has the advantage of tracking the maximum power point, even in the presence of partial shading [65]. An innovative method also allows the performance of PV modules to be measured individually, and their characteristic curves can then be reconstructed [66]. However, one of the main drawbacks of this technique is the adequate sizing of the capacitor, which is necessary to obtain more accurate, regular, and smooth results. Indeed, a capacitor that is too small charges too quickly, distorting the data used to establish the I–V and P–V characteristic curves, while one that is too large slows down the operation. Furthermore, once the capacitor is charged, it must be discharged to repeat the measurement, making the method incompatible with repeated or continuous analyses. To overcome this limitation, the use of dynamic capacitive loads has been introduced (see Figure 19), as demonstrated by the work of Chen, et al. [63,67].

The current and voltage values were measured using dedicated sensors. At the start of the measurement, the activation of the relays created a low-load condition on the PV generator, which was equivalent to a short circuit. As the capacitor charges, the current decreases while the voltage increases. At the end of the charge, the current becomes zero, corresponding to the open-circuit condition.

This technique is capable of operating in various environmental conditions through an automatic scanning system using intelligent, portable I–V and P–V curve plotters, and is economic and scalable for outdoor characterization of the performance of PV modules (see Figure 20) [68].

In the system developed by Sayyad, et al. [68], the pyranometric sensor SP-110-SS is used to measure solar irradiance (G), while the temperature sensors DS18B20 or PT100 are used to record the module temperature (T_m). A capacitor bank, designed with different capacitance values, provides the system with flexibility: a low capacitance is used for small modules, while a higher capacitance is reserved for plotting the curves of the entire system.

In addition, under conditions of low irradiance, the current generated is lower than that under high irradiance. All data is stored in a data logger for further processing.

Finally, in addition to the challenge of correctly sizing the capacitor, another disadvantage of this method is the impossibility of reproducing the I–V curve cyclically. Thus, direct visualization or partial reproduction of this curve remains limited.

Inductive Charge Characterization

Borekci et al. [69] developed a simple, fast, and efficient technique for the analysis of the I–V curve of a PV module using an inductor as a charge (see Figure 21). The proposed method exhibits a reduced response time and increased accuracy. A power MOSFET of the IRF740 type is used to connect the inductor in series with the PV module. In parallel to the inductor, a high-speed diode (MUR1560) and an adjustable discharge rheostat are installed to limit overvoltage on the switch during operation.

During the tests, the voltage and output current of the PV system were measured using an oscilloscope. When the MOSFET is activated, the module quickly switches from an open circuit to a short circuit, generating instantaneous current and voltage values. These variations are shown in Figure 22, corresponding to sunshine conditions of 1000 W/m² and 800 W/m², respectively, at an ambient temperature of 29 °C.

Characterization by DC–DC Converter

DC–DC converters are also used for the characterization of PV modules by allowing the variation of the virtual resistance at their input terminals, thus serving as I–V curve tracers [70,71,72]. An example of this principle is illustrated by the system proposed in Figure 23 [73].

Commonly used types of DC–DC converters include the Down–Boost, Cuk, Zeta, and Asymmetric Primary Inductance (SEPIC) (see Figure 24) [26,74]. The converter is typically inserted between the PV module and the load to allow maximum power point (MPP) tracking (see Figure 25) [75].

The use of DC–DC converters makes it possible to design versatile measuring systems capable of testing and following the evolution of the I–V curves of individual modules or complete photovoltaic fields [73]. In this regard, Durán et al. [76] developed a method using a SEPIC converter capable of capturing and displaying the I–V and P–V characteristic curves of a PV module in real time. This approach takes into account environmental parameters such as irradiance and temperature, which are measured using a calibrated pyranometer and Pt100 sensor, respectively, located at the back of the PV module.

This technique has several advantages over conventional methods, including a simple structure, fast response, high versatility, real-time display, and reduced cost [72]. It also enables reliable and rapid measurement of characteristic curves, with low ripple due to variations in external conditions such as solar irradiation.

However, the method based on DC–DC converters also has some constraints. It requires complex programming of the control signal, and the main design challenges are managing the response speed and power dissipation. To achieve their optimal operation, a filter capacitance greater than the maximum value of the minimum filter capacitance of the converter is required. This is performed in order to limit the output voltage ripple. They have a high power loss in the diodes due to their configuration. They generally have a higher power loss in energy storage elements, such as capacitors and inductors, as the input current increases [77,78]. High-frequency switching operations cause ripples and increase the difficulty of smooth data acquisition [26].

2.2.2. Characterization of the Effects of Environmental Parameters

Characterization of the Effect of Sunlight

The main characteristics of solar irradiation incident on PV modules are power, spectral distribution, and angle of incidence [79]. At a constant temperature, the short-circuit current increases proportionally to the incident illumination level, while the open-circuit voltage increases logarithmically [80]. Therefore, illumination strongly influences the current generated by the PV module, resulting in an increase in the power delivered and thus an increase in the maximum power of the PV module [81,82,83].

Numerous studies have explored the impact of sunlight on the performance of PV modules using various approaches. For example, Chukwu et al. [84] conducted a real-world analysis in Lagos (Nigeria), a city with a hot and humid climate, and compared the performance of four commercial PV modules. Parameters such as output power, module efficiency, and performance ratio were recorded for three consecutive days from 7:30 a.m. to 7:30 p.m. using a data logger. Solar irradiation, measured hourly by a pyranometer at the same inclination as the PV modules, has been shown to have a linear effect on output power. Ibrahim, et al. [85] collected 28 sets of experimental data over 14 days and analyzed them using Microsoft Excel. Their results showed that low solar radiation significantly reduced the power produced by the PV system. Similarly, Buni, et al. [86] showed a direct relationship between solar radiations, current, voltage, and efficiency of the solar panel. Their numerical measurements confirmed that an increase in irradiation led to an increase in the output current and, therefore, the efficiency of the module. Ezenwora, et al. [87] evaluated three silicon PV module technologies, using a Li-200SA M200 pyranometer coupled with a CR1000 (Campbell Scientific, Inc., Lincoln, NE, USA) data collection system. The measurements were carried out daily from 8 am to 6 pm for a whole year. Their results indicated a positive influence of solar radiation on the energy efficiency of the modules. At the National Solar Energy Institute in Gurgaon-Faridabad (India), Siddiqui et al. [33] conducted outdoor measurements over ten months in a high-irradiation and temperature climate. Using a modern test system (I–V and P–V characterizations), they showed that the performance of the modules decreased during cold months due to low cumulative radiation and improved in warmer months. Bashir, et al. [88] conducted a study in Taxila (Pakistan) during the winter, analyzing the efficiency and performance ratio according to solar irradiation measured with a TBQ-2 pyranometer. The data, collected every hour from 8 a.m. to 5 p.m., on alternate days based on the time when the sun appears, revealed an improvement in performance with increasing irradiation, and a marked decrease in the case of a decrease in irradiation. In Romania, Cotfas, et al. [89] compared two PV module technologies (m–Si and a–Si) in Brasov over a period of two years. Using an SPN1 pyranometer, the data were recorded every 5 min. Results indicate that under low light, the maximum power generated by the monocrystalline module can be up to 2.4 times higher than that of the amorphous module. Al–Bashir, et al. [90], proposed an analytical model based on multivariate linear regression to estimate the power generated by a PV system from solar irradiation data. The measurements, collected every minute between September 2016 and January 2017, show that solar radiation has a more significant influence on output power than other factors such as temperature or wind speed. Islam et al. [91] experimentally demonstrated that the maximum power of modules increases with increasing solar irradiation. A pyranometer was used to measure the irradiation, and curves (I–V) and (P–V) were obtained using a potentiometer adjusting current and voltage. Finally, Baghel, et al. [92] conducted a comparative study of mono- and polycrystalline modules in wet and dry tropical climates in India from October 2020 to March 2021. The measurements, carried out with a pyranometer and recorded every 20 s, showed that the output power and efficiency increased with irradiation. However, above a certain threshold, the performance decreased due to temperatures above the STC conditions, highlighting the negative effect of overheating (see Figure 26).

In summary, as shown in

Table 2. Summary table of the effect of solar radiation on the performance of PV modules.

Authors	Year	PV Technology	Exposition Period	Module Efficiency	Performance Ratio	Climatic Conditions
CHUKWU, et al. [84]	2016	m–Si	Three day	11.83%	0.900	hot and humid
		p–Si		9.16%	0.923
		CIS		7.12%	1.065
		a–Si		3.61%	1.023
Ezenwora, et al. [87]	2018	m–Si	one year	5.86%	0.07	tropical
		p–Si		10.91%	0.13
		a–Si		3.61%	0.07
Bashir, et al. [88]	2014	m–Si	winter	7.8%	5.68%	semi-arid
		p–Si		8.2%	9.3%
		a–Si		22.1%	22.6%
Cotfas, et al. [89]	2019	m–Si	2 years	8.34%	-	continental
		a–Si		3.48%
Islam, et al. [91]	2014	m–Si	-	13.6%	-	tropical
Baghel, et al. [92]	2022	m–Si	6 month	-	0.89	humid tropical
		p–Si			0.86

, the results of the effect of solar radiation on the performance of PV modules vary from one area to another depending on the climatic conditions in which the studies were carried out and the amount of solar irradiation falling on the material.

Characterization of the Temperature Effect

The temperature variation of a PV module significantly influences its electrical performance, including short-circuit current, maximum power, open-circuit voltage, form factor, and overall efficiency [81,83]. This parameter is a major issue because it is closely linked to the energy efficiency of the system [93].

A characterization study of the impact of temperature on PV module performance was conducted by Jatoi et al. [94,95] in the subtropical desert climate of Nawabshah, Pakistan. The authors evaluated the performance of four PV module technologies (p–Si, m–Si, a–Si, and thin-film) based on their efficiency. The surface and backside temperatures of each module were recorded using K-type thermocouples (models Prova-210 and Prova-830). Similarly, Jaszczur et al. [93] conducted a similar study in a temperate climate on crystalline modules (p-Si, m-Si) installed at the University of Science and Technology in Cracovie, Poland. For the latter, the thermal distribution within the PV modules was studied by installing eight T-type thermocouples at the back of two modules (four per module) (see Figure 27). Regardless of the climate, the results of the two studies showed a progressive decrease in the efficiency of the modules as the temperature increased. However, this decrease is more marked for crystalline technologies (p–Si, m–Si, a–Si) than for thin-film modules, suggesting a better thermal tolerance for the latter. Considering the climates of the two study sites, we noticed a decrease in the efficiency of 0.40%/°C, 0.34%/°C, 0.29%/°C, and 0.25%/°C for p-Si, m-Si, a-Si, and thin films, respectively, in the climate of Nawabshah, and 0.42%/°C and 0.38%/°C, respectively, for p-Si and m-Si in the climate of Cracovie. These results indicate that the climate of Nawabshah is much more favorable than that of Cracovie for these two technologies.

Takyi, et al. [96] studied the impact of temperature on the performance of a monocrystalline silicon PV module under the climatic conditions of Kumasi, Ghana, using a specific cooling method. The module was first cooled in an insulated chamber to a temperature between 10 °C and 15 °C. Cooling the PV module at low temperatures allowed different I–V curves to be plotted and analyzed over a wider temperature range. This made it possible to obtain temperature coefficients at different temperatures from the plotted I–V curves. Before being subjected to outdoor electrical tests, the module was first covered with cardboard to maintain the temperature imposed on it until the start of the test. It was then quickly removed from the chamber for I–V testing, in accordance with IEC 60904-1. A thermocouple is placed on the back of the module to record the temperature. The experiment was conducted during a period of relatively high radiation. A DS-1000 I–V curve plotter, manufactured by DAYSTAR (Bedford, TX, USA), combined with a laptop computer, was used to obtain an I–V curve by varying the electrical impedance connected to the output terminals of the PV module. Changing the impedance from zero to infinity shifts the module’s operating point from Isc to Voc. The DS-1000 achieves an impedance change by connecting the module to a capacitive load.

The results showed that an increase in temperature induced a decrease in irradiance, thus leading to a decrease in maximum power (Pmax). Furthermore, the open-circuit voltage (Voc) decreases with increasing temperature, while the short-circuit current (Isc) increases slightly.

Al-Ghezi, et al. [97] conducted a study combining numerical and experimental approaches to assess the impact of temperature on the performance of polycrystalline PV modules in the climatic conditions of Baghdad, Iraq. The analysis focused on two key indicators: output power and module efficiency. The measurements were carried out from 8:00 a.m. to 5:00 p.m., with data recorded every 15 min. The electrical output parameters were measured using a Proskit MT-1210 digital multimeter(EMIN GROUP, Hai Phong City, Viet Nam). To monitor the temperature of the module, three K-type thermal sensors covering a measuring range of −200 to 1260 °C were installed on the back of the module at the top, center, and bottom to determine the average operating temperature. Ambient temperature was measured in the shade using a conventional thermometer. According to these authors, the experimental results were compared with numerical simulations performed using the PVsyst software, revealing good agreement between the two data sets. The observations showed that a 1 °C increase in PV module temperature results in an output power decrease of 0.489% and an efficiency reduction of approximately 0.586%. The study shows that the actual performance of PV modules can diverge significantly from standard test conditions (STC) due to weather variations, such as temperature and solar radiation, which directly influence their behavior under real-life conditions.

Fezzani, et al. [98] evaluated the energy efficiency and performance ratio of four different PV module technologies installed at the Renewable Energy Applied Research Unit (URAER) in Ghardaïa, Algeria, a region characterized by a hot and dry climate in summer, and cold in winter. This study was conducted over a 12-month period in accordance with the recommendations of the International Energy Agency (IEA) to provide a representative annual performance assessment. Among the key parameters analyzed was the PV module temperature, which is known to have a decisive influence on the efficiency of photovoltaic systems. This was measured using a Pt-100 sensor (Kipp & Zonen, Delft, The Netherlands), and the data were recorded in real time via an Agilent 34970A acquisition system (Agilent technologies Inc., Lovelend manufacturing Center, Loveland, CO, USA) connected to a computer, ensuring continuous collection and storage of measurements. The results showed that an increase in temperature reduced the energy efficiency of all the module technologies tested. However, this effect was found to be more pronounced for polycrystalline silicon (p–Si) modules than for thin-film modules, highlighting the higher thermal sensitivity of p-Si technologies.

Biodun, et al. [99] conducted an experimental study to assess the effect of temperature on the performance of m-Si and p-Si PV modules in the city of Ogbomoso, located in the state of Oyo, Nigeria. The analysis focused on a key performance indicator: module output power. Measurements were made using digital multimeters to record the temperature and open-circuit voltage, and an AC/DC clamp meter was used to measure the short-circuit current. The data were collected every 30 min, between 9:00 a.m. and 3:00 p.m., over a period of one month, in order to ensure good accuracy and a faithful representation of daily variations. The results showed that between 11:00 a.m. and 1:30 p.m., the increase in temperature was correlated with an increase in output power. However, beyond this time range, a decrease in power was observed, suggesting that overheating of the module above a certain thermal threshold adversely affects its performance. These observations confirm the negative impact of high temperatures on the overall performance of photovoltaic systems.

Ale et al. [100] conducted an experimental study to analyze the influence of temperature on the performance of an m–Si PV module installed at the University of Ibadan, Nigeria. The experiment was conducted over a period of four months, with a configuration designed to record the module’s thermal and electrical data at regular intervals. Measurements were made every hour between 8:00 a.m. and 6:00 p.m., using K-type thermocouples to record the temperature of the module and multimeters to measure voltage and current. The results showed a positive correlation between the module temperature and output power, as well as the efficiency. In other words, the increase in temperature was accompanied by an improvement in the electrical performance of the module. However, this behavior is a departure from the trends generally reported in the literature, where an increase in temperature is usually associated with a deterioration in performance. This divergence could be attributed to experimental or environmental conditions specific to the study site, requiring additional investigations to determine the exact causes.

Al-Odat [101] conducted an experimental study to assess the effect of water cooling on the performance of PV modules in the summer weather conditions of the city of Irbid, Jordan. Two identical p–Si PV modules were used: one with a water heat exchanger cooling system and the other as a reference PV module. The cooling system consisted of a tank that maintained an almost constant water temperature, with circulation through a brass pipe attached to the back of the cooled PV module. The water absorbed the heat released by the module before being directed to a domestic hot water production device, thus integrating thermal recovery. The experiment was conducted during the summer from 8:00 a.m. to 6:00 p.m., with measurements taken every hour on both PV modules under identical environmental conditions. The front and back surface temperatures of the ambient air and those of the inlet and outlet water were recorded using infrared thermometers, and the electrical parameters (voltage and current) were measured using digital multimeters. The results confirmed that the temperature of the PV module is a critical factor that affects its energy performance. The cooling system reduced the surface temperature by approximately 20%, resulting in an improvement in the electrical conversion efficiency of about 14%. These results highlight the potential of photovoltaic-thermal (PV/T) hybrid systems for optimizing the energy output of solar installations, especially in hot climates.

Onyenweuwa [102] conducted a comparative study to analyze the effect of temperature on the electrical performance of two m-Si PV modules, one with an active cooling system and the other without thermal control (see Figure 28). The measured parameters included ambient temperature, module temperature, and output voltage of both cooled and uncooled modules. These data were recorded using thermometers, thermostats, and voltmeters over a period of 8 h from 9:00 a.m. to 5:00 p.m. The module efficiency was evaluated using the calculation method proposed by Lee, et al. [103]. The results showed that an increase in PV module temperature led to a decrease in voltage and yield, while a decrease in temperature had the opposite effect. The PV modules with active cooling showed higher voltages and efficiencies than the uncooled PV modules, demonstrating that active thermal management can significantly improve the performance of photovoltaic systems.

Continuing work on thermal impact, Ale, et al. [104] applied a water-cooled technique at the Faculty of Technology of the University of Ibadan, in the state of Oyo (Nigeria), to evaluate the performance of PV modules in tropical climates. A PV module was installed outdoors, and measurements were taken between October 2017 and January 2018. Module and ambient temperatures were measured using an infrared sensor and a K-type thermocouple, respectively. Two separate multimeters were used to record voltage and current, allowing the power generated to be calculated. Data were collected every hour from 8:00 a.m. to 6:00 p.m. for two days. The same procedure was repeated with the integration of a water-cooling system, using a 0.5 HP pump to spray water on the surface of the PV module. The results showed that under cooling, the temperature of the PV module decreased by 24.1 °C, resulting in a 10.06% increase in the open-circuit voltage, a 9.83% increase in the power delivered, and a 9.83% improvement in the conversion efficiency. The cooling system stabilized the PV module temperature in the range of 30.6 °C to 34.5 °C.

Yuldoshov, et al. [105] conducted an experiment to analyze the effect of temperature on the electrical performance of a PV module under real-world conditions at the Termez State University site in Uzbekistan. The experiment was conducted on October 4, 2022, between 8:00 a.m. and 4:00 p.m. using two identical m-Si PV modules. K-type thermocouples were used to measure temperatures at different points of the module: the front glass, the back side, and the back electrical contact of the first PV module. At the same time, the open-circuit voltage (Voc) and short-circuit current (Isc) were measured using digital multimeters. The measurements were carried out at an ambient temperature of 30 °C and solar radiation intensity of 850–950 W/m². The results revealed a significant temperature variation between the different layers of the module: a difference of 20 °C between the front glass and the back sheet, and 25 °C between the back electrical contacts. This thermal mismatch caused the open-circuit voltage to drop from 21 to 19.3 V, while the short-circuit current increased slightly from 0.65 to 0.75 A. These variations confirm the direct influence of temperature on the key electrical parameters of PV modules. This causes a loss of approximately 35% of the electrical energy. This thermal mismatch gradually creates a hot spot between PV module contacts, further degrading the reliability of the PV module. This problem can be reduced or avoided by placing a thermal insulator between the different module contacts.

Characterization of the Dust Effect

Dust comprises very fine particles, usually less than 500 μm in diameter. Its composition and concentration vary according to the nature of the soil, climatic conditions, and level of human activity in each locality. Numerous experimental studies have shown that dust has a significant negative effect on the performance of PV modules. In fact, it can lead to a loss of up to 50% in some cases of PV plants exposed to significant accumulation [106]. In order to better understand and quantify this influence, several studies have been conducted, which are presented below.

In this context, Lakshmi et al. [107] conducted an experimental study on the impact of different categories of dust (chalk, brick, coal, and sand) on the performance of four PV modules exposed to four distinct levels of solar irradiation under a clear sky at the R.M.K. Engineering College, Chennai, India. The objective of this study was to evaluate the performance of PV modules in a tropical environment that is highly exposed to dust particles. Dust samples were collected from various environments (industrial areas, agricultural land, etc.) and applied uniformly to the PV modules in equal quantities. After application, the PV modules were left to rest for 30 min to allow natural decantation of the particles. The comparative performance of clean and dirty modules was evaluated through output power and efficiency measurements. The instrumentation included an ammeter and a digital voltmeter (for current and voltage), a variable resistive load to plot curves (I–V), a Solar Power meter SM206 for irradiation, a precision balance for the weight of dust, and a multimeter for various electrical measurements. The results showed that the ratio of the maximum power of the dirty module to that of the clean module varied between 26.49% and 90.44%, depending on the type of dust. The maximum observed efficiency losses are 73.51% for coal, 66.29% for sand, 65.46% for brick powder, and 61.42% for chalk. It is shown that coal dust is the most harmful in terms of performance, due to its high absorption capacity and minimal light transmissivity.

Another experimental study conducted by Andrea et al. [108] aimed to evaluate the sensitivity of a p–Si PV module to industrial dust deposition, depending on the type and particle size, in the tropical climate of Arusha, Tanzania. The tests were carried out outdoors under different solar irradiations (720 W/m², 800 W/m², and 900 W/m²). Two identical PV modules were used: one was kept clean as a reference, and the other was covered with dust for comparative evaluation. Four types of industrial dust were selected from the fertilizer, gypsum, aggregate crushing, and coal mining industries. The samples were sieved by particle size analysis into three particle size classes: 90–180 μm, 45–90 μm, and 20–45 μm. For each test, 10 g of dust was evenly distributed on the PV module. At the end of each test, the soiled PV module was cleaned with a clean brush to ensure the accuracy of subsequent measurements. Currents and voltages were measured using digital multimeters, and solar irradiation was measured using a Solar Power Meter. A rheostat was used to plot the I–V and P–V characteristics of clean and dusty PV modules. According to the results obtained, the dust type greatly influences the energy efficiency of PV modules. The maximum efficiency losses observed are 64% for coal dust, 42% for aggregate dust, 30% for gypsum dust, and 29% for organic fertilizer dust. These results confirm that coal dust is the most harmful, due to its high absorption capacity and, therefore, its very low light transmissivity. In addition, the I–V and P–V curves obtained at different irradiations showed a marked negative impact on the performance of PV modules in the presence of dust.

Rao et al. [109] conducted a dynamic study on performance losses due to dust accumulation by analyzing the I–V characteristics of PV modules. The experiment was carried out using a test bench on the roof of the Centre for Sustainable Technologies at the Indian Institute of Science in Bangalore. Two identical p-Si PV modules were used. I–V characteristics were plotted and recorded using solar module analyzers. Initially, the two PV modules showed a similar level of dust deposition, and the results showed that their I–V curves were almost identical, indicating equivalent electrical behavior under homogeneous dirt conditions. One of the PV modules was cleaned, while the other was kept dusty. In order to better analyze the evolution of their performance as a function of solar irradiation, the I–V characteristics of the two PV modules were recorded at an interval of five minutes. The results clearly show that cleaning the PV modules significantly improves their electrical performance, confirming the negative impact of dust on the efficiency of photovoltaic systems.

Rashid, et al. [110] conducted a study to explore the effects of dust accumulation on the energy production of p–Si PV modules in two distinct climate regions of Pakistan: Islamabad and Bahawalpur. In each region, a PV module was exposed to an outdoor environment for six weeks, in order to compare its efficiency with that of a regularly cleaned module. At the same time, in order to determine the particle size and elemental composition, dust samples were collected from both cities. This was done to understand the impact of dust on the energy efficiency of the PV module. A progressive reduction in the output power (see Figure 29) and system efficiency (see Figure 30) was observed in the results obtained. This reduction was proportional to the number of weeks of exposure in both regions. After six weeks of exposure, the reduction in output power was 15.08% in Islamabad, while in Bahawalpur, it was 25.42%, which is approximately 1.5 times higher. These results indicate that the accumulation of dust has a greater impact in regions with drier and dustier weather conditions, such as Bahawalpur.

Chen, et al. [111] studied the influence of the density of deposited dust on PV modules on their performance in Hong Kong, China. The study involved the use of four p–Si PV modules, one of which was kept clean, and the other three were exposed to different densities of dust (0 g/m², 10 g/m², and 30 g/m²). The dust was evenly distributed over the surfaces of the three PV modules. A weather station and data logger were used to measure and record the solar radiation, ambient temperature, and PV module temperature. In addition, a PV module analyzer was used to measure the voltages, currents, and powers of the four PV modules. The results showed that the conversion efficiency of the modules and form factor (FF) decreased proportionally with increasing dust density. In addition, a non-linear correlation was observed between the dust density and conversion efficiency (see Figure 31). These results are consistent with those obtained by Rashid, et al. [110], confirming the significant impact of dust on photovoltaic module performance in environments exposed to high levels of dust.

Ali, et al. [112] investigated the effects of dust accumulation on the performance of m–Si and p–Si PV modules. The experiment was carried out under real conditions in Taxila (Pakistan) during three months of winter, a period propitious to the observation of a progressive deposition of dust.

Four PV modules were tested, of which one of each type was kept clean for comparison purposes. Measurements were made for different dust densities, with I–V data recorded twice a week on sunny days and three times a day (at 9:00 a.m., 12:00 p.m., and 3:00 p.m.). The voltage and current of each PV module were measured simultaneously using digital multimeters.

The results showed a gradual decline in the performance of PV modules with an increase in the amount of dust deposited on their surface. At a density of 0.9867 mg/cm², the average power output decreased by 20% for m–Si modules and 16% for p-Si PV modules. The overall efficiency of the PV modules also suffered significant degradation due to dust deposition, confirming the negative impact of particulate pollution on the performance of photovoltaic systems.

Adinoyi et al. [113] conducted an in-depth study on the impact of dust accumulation on energy production from six outdoor PV modules exposed in Dhahran, Saudi Arabia. The objective was to evaluate the degradation of PV module performance according to the level of fouling over a prolonged period. The output power was measured daily using an HT Italia IV-400 PV analyzer. The measurements included PV module currents and voltages, ambient temperature, module back temperature, and overall radiation incident on the inclined surface. Dust samples were collected, weighed to determine their density, and analyzed using energy dispersion spectroscopy (EDS) to determine their elemental composition. EDS analysis revealed that oxygen was the major component of the dust particles. The results showed that a decrease in output power of up to 50% could be observed for modules left uncleaned for more than six months. These results highlight the significant impact of prolonged particulate pollution on the energy efficiency of photovoltaic systems, especially in desert or semi-arid environments.

Rahman, et al. [114] examined the impact of dust accumulation on the performance of an m–Si PV module, installed outdoors for one month. This study aimed to compare the performance of clean and dust-exposed modules by analyzing their electrical characteristics. The current and voltage of both modules (clean and dirty) were measured using a digital multimeter. The results showed that the short-circuit current of a clean PV module was consistently higher than that of a dust-covered module. Given that the output power of a solar module is directly related to this parameter, the maximum power point (MPP) decreases gradually as dust accumulates. According to the authors, the reduction in output power can reach up to 20%, confirming that dirt is a significant factor in the degradation of PV module performance in outdoor environments.

Danu, et al. [115] have carried out an experiment to evaluate the influence of dust deposition on the performance of PV modules, within the Passive House Laboratory of the Faculty of Energy Engineering of the Polytechnic University of Bucharest. The experiment involved covering a PV module with varying amounts of dust while maintaining near-constant lighting and temperature conditions to isolate the effect of dust. The results revealed that the decrease in the maximum power point (MPP) was directly proportional to the amount of dust deposited on the surface of the PV module. More generally, the authors point out that the reduction in output power due to fouling depends not only on the exposure time of the module, but also on the frequency and intensity of dust deposition, which vary according to local environmental conditions. Therefore, it is recommended to implement regular cleaning protocols for PV modules, adapted to the geographical area and its climatic specificities, in order to optimize the system’s performance over time.

In summary, the impact of dust on PV module efficiency varies depending on its composition, density, local climate, and PV technology used. Indeed, dark dusts such as coal or sand cause the highest losses (up to 73.5%), while lighter industrial dusts have a moderate effect (29–30%); desert regions, with prolonged accumulation, cause drops above 50%, while humid tropical climates cause moderate losses despite rapid accumulation; monocrystalline (m–Si) modules are slightly more sensitive than polycrystalline (p–Si), with an efficiency gap of about 4% according to some studies; dust density above 10 g/m² leads to a non-linear performance drop, especially in the form factor (FF); It is therefore recommended to clean the panels frequently, especially in desert areas (every two months to limit losses to less than 20%), to choose the PV technology according to the environmental context, and to use standardized procedures (IEC 60891) to compare performance in real conditions with STC reference data.

Table 3. Comparison of the different efficiency losses of PV modules due to dust and environmental conditions. ↑ indicates whether the dust density increases; ↓ indicates the form factor decreases. these explain the non-linearity.

Authors	Year	Dust Type/Conditions	Dust Density	Efficiency Losses	Climatic Conditions	PV Tested
R, K.; Ramadas, G. [107]	2022	Chalk, brick, coal, sand	Equal application	26.5–73.5%	Tropical (India)	Polycrystalline (p-Si)
Andrea et al. [108]	2019	Industrial dust (fertilizers, gypsum, coal)	10 g/module	29–64%	Tropical (Tanzania)	Polycrystalline (p-Si)
Rashid et al. [110]	2023	Urban dust (Islamabad vs. Bahawalpur)	6-week accumulation	15–25.4%	Semi-arid (Pakistan)	Monocrystalline (m-Si)
Chen et al. [111]	2019	Urban dust (Hong Kong)	0–30 g/m2	Non-linear (↑ density = ↓ FF)	Humid subtropical	Polycrystalline (p-Si)
Ali et al. [112]	2017	Winter dust (Taxila)	0.9867 mg/cm2	16% (p-Si)–20% (m-Si)	Dry winter (Pakistan)	m-Si and p-Si
Adinoyi et al. [113]	2013	Desert dust (Saudi Arabia)	>6 months of accumulation	Up to 50%	Desert (Dhahran)	Unspecified

presents a comparison of the different efficiency losses of PV modules due to dust and environmental conditions.

Characterization of the Effect of Degradation

PV module degradation is one of the key factors to consider in reducing the cost of electricity generated while extending the operational life of PV systems [79]. From the I–V curve, the effects of these degradations due to defects and aging can be observed through the evolution of individual electrical parameters [116]. During exposure, PV modules may undergo degradation, resulting in a gradual change in their performance parameters.

Thus, Lillo-Sanchez et al. [117] conducted a study on the main signs leading to the degradation of crystalline silicon PV modules after 22 years of sun exposure in Seville, Spain. The analysis was carried out by visual inspection, infrared thermography, and electroluminescence (EL) to evaluate the electrical performance of the PV modules. The most notable defects observed included pronounced browning, the appearance of a milky tint, and oxidation of the metal grid. Results showed an average peak power loss of 30.9% over the entire period, or about 1.4% per year. This degradation is mainly attributed to a decrease in the short-circuit current and, to a lesser extent, a decrease in the fill factor and open-circuit voltage (see Figure 32).

Phinikarides, et al. [118] have analyzed a 1 kWp photovoltaic installation installed outdoors at the University of Cyprus, in order to assess the rate of degradation over two successive periods: the first three months of exposure and then the following four months. This study was based on the measurement of the maximum power and system performance ratio, applying a temperature correction to the MPP. The results showed a power loss of 16.9% from baseline in the first three months of operation. Over the next four months, a further 8.6% decline was observed, as shown in Figure 33.

Rajput, et al. [119] conducted a degradation analysis of 90 single-crystal silicon PV modules installed on the roof of the National Solar Energy Institute (NSEI) guest house in Gurgaon, after 22 years of outdoor operation in an Indian composite climate. The analysis was carried out by visual inspection, thermal imaging, I–V characterization, and measurement of insulation resistance, thus allowing the calculation of the rate of degradation. The most frequently observed defects were bus bars, cell interconnect tapes, chain interconnect tapes, and backsheet gelling. Hot spots in the solar cells, burn marks, and delamination phenomena on the back sheet were also detected in some PV modules. Electrical parameters such as power output (P_max), short-circuit current (I_sc), open-circuit voltage (V_oc), and fill factor (FF) were evaluated as shown in Figure 34.

Piliougine et al. [120] analyzed the degradation of PV modules to amorphous silicon after 11 years of exposure in southern Spain. The experiment was conducted according to procedure 3 of irradiation and temperature correction of IEC 60891:2021, in order to study the evolution of electrical parameters (currents and voltages) over three distinct periods: during the first days, the first year, and the tenth year of exposure. The I–V curves were measured using a suitable measurement system (PVPM 6020C and Kepco/Agilent) (see Figure 35), then adjusted using the equations in procedure 3 of IEC 60891:2021 [41]. The electrical parameters were compared at different stages of the exposure period under the same operating conditions. An initial significant decrease in potency was observed during the first nine days of exposure. After this initial decline, degradation continues at annual rates slightly above 3% for a-Si PV modules and close to 2.3% for micromorphic PV modules. Over the 10-year period, annual power degradation rates are stable at around 1.0%/year for both types of PV modules. These results are hardly superior to those obtained for crystalline silicon PV modules measured at the same location during the same exposure period. If the first year of exposure is included in the 11–year calculation, the annual degradation for the amorphous PV module is approximately 1.12%/year, while for the micromorphic PV module it is approximately 0.98%/year.

Ishii et al. [121] analyzed the degradation rate of crystalline silicon PV modules in Saga prefecture, Japan. Three indicators—energy efficiency, performance ratio, and output power–were used to estimate the annual degradation rates of individual outdoor PV modules over a 3–year period. The authors observed a 2% decrease in performance due to sun-induced degradation. The annual degradation rate was estimated to be 0.2%. In comparison, the analysis conducted by Obaid, et al. [122] in Baghdad, Iraq, estimated these rates at 0.59% per year.

Rajput et al. [119] conducted a degradation analysis of 90 m–Si PV modules installed on the roof of the National Solar Energy Institute (NISE) guest house in Gurgaon, after 22 years of outdoor operation in a composite climate in India. The analysis was carried out using visual inspection, thermal imaging, I–V characterization, measurement of insulation resistance, and calculation of the degradation rate. The results showed an average degradation rate of about 1.9% per year.

Aboagye et al. [123] conducted a study to quantify degradation rates to predict the lifetime of 16 PV systems using different PV module technologies installed in various locations under outdoor conditions in Ghana. This study presented the frequency distribution of the degradation of the output power of PV modules, all older than 5 years, estimated their power degradation rates, and predicted and compared their lifetimes. The electrical parameters of the PV modules were measured, and I–V curves were plotted using automatic tracers connected to the PV modules. Degradation rates were assessed from the translated I–V curves obtained using the equations used by Quansah et al. [124]. The results showed that the degradation rates of PV modules varied not only from one technology to another [125] due to the different materials and manufacturing processes used, but also according to the study area, as stated by Quansah et al. [126]. PV modules in Ghana degraded at rates faster than the standard guarantee rate (see Figure 36).

Gyamfi, et al. [127] conducted a study on evaluating the power degradation rate of p-Si PV modules and comparing them with those of PV modules without visual defects. These PV modules were installed under climatic conditions similar to those in Kumasi, Ghana. A total of 48 PV modules from 11 different manufacturers installed for between 5 and 9 years were sampled from 12 PV installations. The methods used included visual inspection, I–V curves, and experimental and translated waveforms. The visual inspection data collection tool was a checklist developed by the National Renewable Energy Laboratory (NREL). The results showed that the average degradation rates of PV modules varied among manufacturers, ranging from 0.79% to 1.67% per year.

Sadok et al. [128] presented results concerning the behavior of m-Si PV modules in the Adrar region of southern Algeria. This study assessed the long-term degradation of PV modules using visual inspection. The electrical parameters of the PV modules were determined using a graphical adjustment approach based on the least-squares method. The translation procedure [4] was applied to extrapolate the I–V characteristics adjusted to standard conditions. The results showed an average annual rate of degradation in the PV module power of about 1.5%.

Atia, et al. [129] investigated the degradation of 24 m-Si PV modules installed on the roof of the Egyptian Electronic Research Institute (ERI). These PV modules operate for 25 years. The degradation rates were evaluated using visual inspection, I–V characteristic measurements, and degradation rate calculation methods. The experimental results showed that the environmental conditions in Cairo did not significantly impact the performance of the PV modules. This is particularly important because the PV modules continued to operate efficiently, exceeding their expected lifetime, and the degradation rate after 25 years of operation remains within acceptable values.

Daher, et al. [130] presented an analysis of the performance degradation of a 62.1 kWp solar PV power plant located in Ali Adde (Djibouti), after 9.5 years of operation in hot and desert climate conditions. The I–V and P–V characteristic curves of the entire power plant were used to assess performance degradation. The authors used a PV analyzer called Solmetric to plot the I–V and P–V curves of the PV modules in the power plant. The PV analyzer works with the Solsensor, which measures the PV module temperature (Tm), module plane irradiance (POA), and tilt angle. The measured I–V characteristics were then returned to the standard test conditions (STC) to better interpret the results. The degradation rates of the maximum power and fill factor were found to be 0.84%/year and 0.66%/year, respectively.

Based on the various studies consulted, it appears that the degradation rate of PV modules depends on several factors, including the technology used, the manufacturer, and environmental conditions. As a result, it is difficult to directly transfer the results of degradation analyses from one country to another.

Table 4. Annual degradation rates and degradation mechanisms of photovoltaic (PV) modules.

Authors	Year	PV Technology	Study Period	Annual Degradation Rate	Main Degradation Mechanisms	Climatic Conditions
Lillo-Sanchez et al. [117]	2021	m-Si	22 years	1.4%	Browning, oxidation, discoloration	Mediterranean climate (Spain)
Phinikarides et al. [118]	2012	a-Si	7 months	16.9% (3 months) → 8.6%	Accelerated initial degradation	Hot climate (Cyprus)
Rajput et al. [119]	2016	m-Si	22 years	1.9%	Delamination, hot spots, burns	Mixed climate (India)
Piliougine et al. [120]	2022	a-Si/μ-Si	11 years	1.12% (a-Si), 0.98% (μ-Si)	Light-induced degradation (LID)	Southern Spain
Ishii et al. [121]	2017	m-Si	3 years	0.2%	LID	Temperate climate (Japan)
Aboagye et al. [125]	2021	p-Si	>5 years	0.79–1.67%	Varies depending on technology	Tropical climate (Ghana)
Sadok et al. [128]	2016	p-Si	ND	~1.5%	Delamination, corrosion	Algerian Sahara
Atia et al. [129]	2023	m-Si	25 years	<1%	Low degradation	Desert climate (Egypt)

summarizes the annual degradation rates and degradation mechanisms of the photovoltaic (PV) modules.

2.2.3. Discussion and Comparative Analysis of Characterization Techniques in Real Conditions

The characterization of PV modules under natural conditions is essential for evaluating their actual behavior in the field. This allows the main characteristics of PV modules to be obtained, reflecting the effects of environmental parameters such as solar irradiation, temperature, dust deposition, humidity, shading, degradation, and orientation of the modules. Several techniques have been developed for this purpose, each with its own strengths and limitations. These methods include techniques, resistive load, electronic load, capacitive load, inductive load, and the use of DC–DC converters. The resistive load technique is one of the simplest and most economical techniques. It consists of connecting different variable resistors to the module output to scan the operating points of the short circuit and open circuit. Its main advantage lies in its simplicity of implementation and low cost, making it an accessible solution for educational contexts or basic testing of small generators. However, this method is very sensitive to environmental variations during scanning, especially if it is manual, which leads to low accuracy. It is also poorly suited to high-power systems because adequate resistances are expensive, bulky, and prone to heating. To overcome these limitations, some works have integrated microcontroller automation systems, such as Arduino [57], allowing for faster and reproducible scanning. The electronic charge technique, on the other hand, relies on the use of active components such as MOSFET transistors, which simulate a variable charge controlled electronically. This method offers a fast scan, on the order of 80 ms in some cases, and allows for better tracking of environmental fluctuations during measurement [61]. It is easily automatable at a reduced cost, making its use interesting in laboratories or in low-power PV systems. Nevertheless, it generates significant heat dissipation, limiting its use to moderate-power modules [62,63], and remains sensitive to the quality of the components used. Despite this, it constitutes a relevant compromise between cost, speed, and precision. The capacitive load technique uses a capacitor as a load, which is gradually filled with energy from the module. This method is also simple and inexpensive [64,65,131]. It is effective even in the case of partial shading because the capacitor charges continuously, which smoothes the transient effects. However, the sizing of the capacitor is critical for obtaining a reliable measurement [63,67]. Indeed, a capacitor that is too small charges too quickly, distorting the curve, while a capacitor that is too large slows down the operation. Furthermore, once the capacitor is charged, it must be discharged to repeat the measurement, making the method incompatible with repeated or continuous analyses. Dynamic variants have been proposed to make the process more fluid and reproducible; however, they remain experimental or complex to implement. The inductive load technique involves connecting an inductance to the module, causing a sudden current draw that allows instantaneous capture of current and voltage values. Owing to its speed, it reduces the influence of irradiance variations during measurement. It is suitable for both small and large powers and offers high precision. However, its implementation is delicate, as it can cause dangerous transient surges for the components, requiring the addition of diodes, protection, or regulation circuits [69]. Therefore, it is reserved for experienced users or well-supervised industrial contexts. It also has a moderate cost and high adaptability. Finally, the characterization by DC–DC converter is based on the use of converters (type cuk, Boost, or SEPIC) [26,74] to vary the virtual resistance seen by the module. This allows for an extremely precise and fast scan of the I–V curve, often in real time, while dynamically adapting to variations in the environment. This method is the most advanced and versatile because it allows both characterization and tracking of the maximum power point (MPP). It is suitable for all power levels, including large photovoltaic fields. Its main drawbacks are the programming complexity, high cost of quality components, heat dissipation requirements, and the need for power electronics skills. However, it is essential for modern installations that are connected to monitoring systems or intelligent energy management. A comparative analysis of these techniques shows that there is no universal solution. In terms of precision, inductive load and DC–DC converter techniques are the most efficient, and are capable of quickly and reliably measuring the electrical characteristics of modules. In terms of cost, resistive, electronic, and capacitive loads are more economical but less precise and robust. The complexity of implementation varies greatly, being very low for resistive loads, moderate for capacitive and electronic loads, and high for inductive and converter loads. Regarding adaptability to high power, only techniques with loads, inductive, and DC–DC converters are suitable. In terms of resistance to environmental fluctuations (clouds, shading, wind, and temperature), fast loads like inductive and DC–DC perform well, while slow or manual loads like resistive suffer from high variability. Therefore, the choice of method largely depends on the context. For occasional, simple, and inexpensive requirements, the resistive or capacitive load is sufficient; for more detailed analyses, the electronic or inductive load is preferable; finally, for large installations, the DC–DC method is the most suitable, despite its cost and complexity. The evolution of technologies, particularly with the arrival of microcontrollers, embedded systems, and artificial intelligence, has gradually allowed certain limitations to be overcome, notably by automating control, better managing thermal dissipation, or dynamically adapting the settings according to weather conditions. The comparison of different I–V characterization techniques for PV modules under real conditions is given in

Table 5. Comparison of different I–V characterization techniques for PV modules under real conditions.

Method	Accuracy	Cost	Complexity	Measuring Speed	Adapted Powers	Sensitivity to Variations
Resistive load	Low	Very low	Very low	Low	Low	Very high
Electronic Load	Medium	Low	Medium	fast	Low	Medium
Capacitive load	Medium	Low	Low	Medium	Medium to high	Medium
Inductive charging	high	Medium	high	very fast	Medium to high	Low
DC-DC converter	high	high	Very high	Very high	all	Low

.

In conclusion, with this critical analysis, it can be said that the future of outdoor characterization relies on hybrid and intelligent solutions, combining several methods to benefit from their respective advantages while limiting their weaknesses. This paves the way for finer optimization of solar systems, increased reliability, and more sustainable energy management.

3. Translation Techniques

3.1. I–V Curve Translation Techniques of PV Modules

The major disadvantage of laboratory characterization is the high cost of solar simulators. On-site characterization under natural conditions is a more accessible and less costly alternative. However, the lighting and temperature conditions are no longer controlled, requiring the use of translation equations to adjust the results to Standard Test Conditions (STC). Several translation methods have been developed in laboratories around the world, and some have been incorporated into international standards. The main translation methods are discussed in Section 4, after a review of the characterization techniques in laboratory and natural conditions, presented in the following sections.

In the literature, several works have proposed methods and techniques for translation that allow the estimation of the gap between the experimental I–V and P–V curves and those corresponding to the standard test conditions of PV modules. These techniques range from the Sandstorm method [132], published in 1967, to modern numerical approaches, through the “first procedure” of IEC 60891 [133]. This first procedure describes a system of equations for current and voltage, which mainly applies to irradiation corrections within 40% of the standard irradiation (STC). It is based on an essentially empirical method, as illustrated by Equations (2) and (3). Equation (2) is applicable only to the I–V and P–V curves measured at constant irradiance. In the case of variation in irradiation during measurement, a corrected version of the translated current is given by Equation (4) [41]. This method works well over a wide range of irradiations, but it incompletely reproduces the I–V and P–V curves. For temperature correction, it remains valid as long as the measured irradiation is within 30% of the irradiation at which the temperature coefficients were determined [41].

The second procedure of IEC 60891 is used for various temperature and irradiation conditions. It is based on a semi-empirical method (Equations (5) and (6)) based on the single-diode model (see Figure 37), which is widely used to describe the operation of a cell, module, or PV field. This procedure produces complete I–V curves for higher irradiation correction. This method is particularly suitable when the measured irradiation is within a range of 30% of the irradiation at which the temperature coefficients are defined. However, this method requires knowledge of more parameters than procedures 1 and 4.

The third procedure of IEC 60891 is based on linear interpolation (Equations (7) and (8)) of the measured characteristics (I–V) and (P–V). The pairs of measuring points on the curves (I–V), (I₁, V₁), and (I₂, V₂) must respect the relation defined by Equation (9). The translations in this method are based on two assumptions: translation in light and temperature (Equations (10) and (11)) [2,134]. This method offers better accuracy for irradiation and temperature corrections, but does not allow for a comparison of the performance between different PV module technologies.

The fourth procedure is more suitable for irradiation corrections when experimental and translated values of irradiations are between 300 W/m² and 1200 W/m² [135]. Validated in 2021 [136], this procedure is based on a one-diode model. The measured I–V curve must be translated according to the translated temperature and irradiation values by applying Equations (12)–(15).

An evaluation study of the performance of these four procedures was carried out, taking into account environmental factors such as seasonal variation, irradiation level, and temperature [137]. The results show that the second procedure is relatively more efficient. However, experimental tests using real I–V and P–V curves have shown that this procedure has better correction robustness. The authors revealed that the fourth corrected procedure is less efficient under partial shading and short-circuit conditions. This is due to the dynamic determination of the correction coefficients. However, it has a similar or even superior performance to Procedures 1 and 2 in conditions where PV module defects are often difficult to detect, making I–V correction particularly necessary. Therefore, it is a promising correction procedure, especially when the correction coefficients cannot be determined in advance.

Other researchers have proposed comparable methods, such as those of Anderson at the NREL in the United States in 1996 [138], which was enhanced by King, et al. [139] and Tobías et al. [140] (Equations (16)–(22)). ASTM E-1036 (ASTM, 2007) [141] presents test methods covering the electrical performance of PV modules under real site conditions. According to this method, Equations (23)–(26) are used to correct the measurements of the short-circuit current and open-circuit voltage. The method of Blaesser, et al. [142] proposed a correction of the I–V curve measured under irradiation and temperature conditions (G,T) to the new conditions (G₁,T₁) by choosing an arbitrary point (I,V) on the I–V curve, which is then transformed into a point (I₁,V₁). This allows the characteristic equations to be obtained for the determination of the correction parameters via an algebraic resolution (Equations (27)–(31)). In addition, Castañer’s method [143] proposes an approach to quantify the influence of current variation on the open-circuit voltage using the thermal voltage Vth, expressed by Equation (32). The JRC (Joint Research Centre) method presents an alternative correction procedure by discretizing the variables of the equation obtained with the temperature variation (Equations (33) and (36)). Reddy’s method [144] proposes a simplification of the JRC (Joint Research Centre) method (Equations (37) and (38)), using the basic equations to propose a pair of expressions for quantifying short currents circuit and open circuit voltage, with a correction factor of the irradiation curve, as proposed by the Castañer method [143].

Table 6. Summary of translation procedures and their equations.

Authors	Year	Correction Procedures	Corresponding Equations	Equations Numbers
Sandstrom [132]	1967	First proceeding	${\mathrm{I}}_2={\mathrm{I}}_1+{\mathrm{I}}_{\mathrm{s}\mathrm{c}}\left({\displaystyle \frac{{\mathrm{G}}_2}{{\mathrm{G}}_1}}-1\right)+\mathrm{\alpha }\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(2)
Guenounou, et al. [133]	2016	First proceeding	${\mathrm{V}}_2={\mathrm{V}}_1-{\mathrm{R}}_{\mathrm{s}}\left({\mathrm{I}}_2-{\mathrm{I}}_1\right)-\mathrm{\kappa }{\mathrm{I}}_2\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)+\mathrm{\beta }\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(3)
IEC [41]	2021	First corrected procedure	${\mathrm{I}}_2={\mathrm{I}}_1+{\mathrm{I}}_{{\mathrm{s}\mathrm{c}}_1}\left({\displaystyle \frac{{\mathrm{G}}_1^{\mathrm{\prime }}}{{\mathrm{G}}_{{\mathrm{s}\mathrm{c}}_1}}}\right)\times \left({\displaystyle \frac{{\mathrm{G}}_2}{{\mathrm{G}}_1^{\mathrm{\prime }}}}-1\right)+\mathrm{\alpha }\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(4)
IEC [41]	2021	Second procedure	${\mathrm{I}}_2={\mathrm{I}}_1·\left(1+{\mathrm{\alpha }}_{\mathrm{r}\mathrm{e}\mathrm{l}}\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)\right)·\left({\displaystyle \frac{{\mathrm{G}}_2}{{\mathrm{G}}_1}}\right)$	(5)
			${\mathrm{V}}_2={\mathrm{V}}_1-{\mathrm{V}}_{\mathrm{o}{\mathrm{c}}_1}·\left({\mathrm{\beta }}_{\mathrm{r}\mathrm{e}\mathrm{l}}\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)+\mathrm{a}·\mathrm{l}\mathrm{n}\left({\displaystyle \frac{{\mathrm{G}}_2}{{\mathrm{G}}_1}}\right)\right)-{\mathrm{R}}_{\mathrm{s}}^{\mathrm{\prime }}\left({\mathrm{I}}_2-{\mathrm{I}}_1\right)-{\mathrm{\kappa }}^{\mathrm{\prime }}·{\mathrm{I}}_2·\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(6)
Tsuno, et al. [134]	2005	Third procedure	${\mathrm{I}}_3={\mathrm{I}}_1+\mathrm{A}·\left({\mathrm{I}}_2-{\mathrm{I}}_1\right)$	(7)
			${\mathrm{V}}_3={\mathrm{V}}_1+\mathrm{A}·\left({\mathrm{V}}_2-{\mathrm{V}}_1\right)$	(8)
			${\mathrm{I}}_2-{\mathrm{I}}_1={\mathrm{I}}_{\mathrm{s}{\mathrm{c}}_2}-{\mathrm{I}}_{\mathrm{s}{\mathrm{c}}_1}$	(9)
			${\mathrm{G}}_3={\mathrm{G}}_1+\mathrm{A}·\left({\mathrm{G}}_2-{\mathrm{G}}_1\right)$	(10)
			${\mathrm{T}}_3={\mathrm{T}}_1+\mathrm{A}·\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(11)
IEC [41] Hishikawa, et al. [135]	2021	Fourth proceeding	${\mathrm{I}}_1^{\mathrm{\prime }}={\mathrm{I}}_1+{\mathrm{I}}_{{\mathrm{s}\mathrm{c}}_1}\times \left({\displaystyle \frac{{\mathrm{G}}_2}{{\mathrm{G}}_1}}-1\right)$	(12)
			${\mathrm{V}}_1^{\mathrm{\prime }}={\mathrm{V}}_1-{\mathrm{R}}_{\mathrm{s}}\left({\mathrm{I}}_1^{\mathrm{\prime }}-{\mathrm{I}}_1\right)$	(13)
			${\mathrm{I}}_2={\mathrm{I}}_1^{\mathrm{\prime }}+{\mathrm{\alpha }}_{\mathrm{r}\mathrm{e}\mathrm{l}}\times {\mathrm{I}}_{\mathrm{S}\mathrm{C},\mathrm{S}\mathrm{T}\mathrm{C}}\times \left({\mathrm{T}}_2-{\mathrm{T}}_1\right)$	(14)
			${\mathrm{V}}_2={\mathrm{V}}_1^{\mathrm{\prime }}+\left({\mathrm{T}}_2-{\mathrm{T}}_1\right)\times {\displaystyle \frac{1}{{\mathrm{T}}_1}}\times \left({\mathrm{V}}_1^{\mathrm{\prime }}-{\mathrm{n}}_{\mathrm{S}}\times \mathrm{\epsilon }\right)$	(15)
Anderson [138] King, et al. [139] Tobías, et al. [140]	1996 1997 2010	NREL Procedure	${I_{sc}}_{STC}=\left[{\displaystyle \frac{I_{sc}}{1+{\alpha }_{I_{sc}}\left(T_c-T_{c_{STC}}\right)}}\right]\left({\displaystyle \frac{G_{STC}}{G}}\right)$	(16)
			${V_{oc}}_{STC}={\displaystyle \frac{V_{oc}}{1+{\beta }_{V_{oc}}\left(T_c-T_{c_{STC}}\right)}}$	(17)
			${P_{max}}_{STC}=\left[{\displaystyle \frac{P_{max}}{1+{\gamma }_{P_{max}}\left(T_c-T_{c_{STC}}\right)}}\right]\left({\displaystyle \frac{G_{STC}}{G}}\right)$	(18)
			${I_{max}}_{STC}=\left[{\displaystyle \frac{I_{max}}{1+{\alpha }_{I_{sc}}\left(T_c-T_{c_{STC}}\right)}}\right]\left({\displaystyle \frac{G_{STC}}{G}}\right)=I_{max}\left({\displaystyle \frac{{I_{sc}}_{STC}}{I_{sc}}}\right)$	(19)
			${V_{max}}_{STC}={\displaystyle \frac{V_{oc}}{1+{\beta }_{V_{oc}}\left(T_c-T_{c_{STC}}\right)}}=V_{max}\left({\displaystyle \frac{{V_{oc}}_{STC}}{V_{oc}}}\right)$	(20)
			$F{F}_{STC}={\displaystyle \frac{{P_{max}}_{STC}}{{I_{sc}}_{STC}·{V_{oc}}_{STC}}}$	(21)
			${\eta }_{STC}={\displaystyle \frac{{P_{max}}_{STC}}{G_{STC}·S}}$	(22)
ASTM [141]	2007	ASTM Procedure	$I_{SC}=\left[{\displaystyle \frac{I_{SCU}}{1+\alpha \left(T_C-T_0\right)}}\right]$	(23)
			$V_{OC2}={\displaystyle \frac{V_{OCU}}{\left[1+\beta \left(E_0\right)\times \left(T_C-T_0\right)\right][1+\delta \left(T_C\right)\times ln \left(\frac{E}{E_0}\right)]}}$	(24)
			${\mathrm{I}}_0=\mathrm{I}\times \left({\displaystyle \frac{{\mathrm{I}}_{\mathrm{S}\mathrm{C}}}{{\mathrm{I}}_{\mathrm{S}\mathrm{C}\mathrm{U}}}}\right)$	(25)
			${\mathrm{V}}_0=\mathrm{V}\times \left({\displaystyle \frac{{\mathrm{V}}_{\mathrm{O}\mathrm{C}}}{{\mathrm{V}}_{\mathrm{O}\mathrm{C}\mathrm{U}}}}\right)$	(26)
Blaesser, et al. [142]	1988	Blaesser procedure	$I_1=I+I_{SC}\left({\displaystyle \frac{G_1}{G}}-1\right)$+$\alpha \left(T_1-T\right)$	(27)
			$V_1=V-R_S\times \left(I_1-I\right)$+$\beta \times \left(T_1-T\right)$	(28)
			$I_{STC}=I+I_{SC}\left({\displaystyle \frac{G_{STC}}{G}}\right)-1$	(29)
			$V_{STC}=V+\beta \times (25-T$)	(30)
			$V_{OC,STC}=V_{OC}+\beta \times \left(25-T\right)+N_S\times n\times V_{th}\times ln \left({\displaystyle \frac{G_{STC}}{G}}\right)$	(31)
Lun, et al. [143]	2013	Castañer procedure	$V_{OC2}=V_{OC1}+\beta \times \left(T_2-T_1\right)+N_S\times n\times V_{th}\times ln \left({\displaystyle \frac{I_{SC2}}{I_{SC1}}}\right)$	(32)
Joint Research Centre	--	JRC Procedure	$I_{SC2}=I_{SC1}\times 1+{\alpha }_r\left(T_2-T_1\right)\times \left({\displaystyle \frac{G_2}{G_1}}\right)$	(33)
			$V_{OC2}=V_{OC1}\times (1+{\beta }_r\times \left(T_2-T_1\right)+\delta \times ln \left({\displaystyle \frac{G_2}{G_1}}\right)$	(34)
			${\mathrm{I}}_2^{\left(j\right)}={\mathrm{I}}_1^{\left(j\right)}\times \left({\displaystyle \frac{{\mathrm{I}}_{\mathrm{S}\mathrm{C}2}}{{\mathrm{I}}_{\mathrm{S}\mathrm{C}1}}}\right)$	(35)
			${\mathrm{V}}_2^{\left(j\right)}={\mathrm{V}}_1^{\left(j\right)}+\left(V_{OC2}-V_{OC1}\right)+R_S\times (I_1^{\left(j\right)}-I_2^{\left(j\right)})$	(36)
Reddy, et al. [144]	2017	Reddy’s procedure	$I_{SC2}=I_{SC1}·\left(1+{\alpha }_r\left(T_2-T_1\right)\right)·\left({\displaystyle \frac{G_2}{G_1}}\right)$	(37)
			$V_{OC2}=V_{OC1}\times (1+{\beta }_r\times \left(T_2-T_1\right)+\delta \times ln \left({\displaystyle \frac{G_2}{G_1}}\right))$	(38)

summarizes the different translation procedures and associated equations.

3.2. Discussion and Comparative Analysis of I–V Curve Translation Techniques of PV Modules

The techniques for translating the I–V curves of photovoltaic modules aim to bring back the measurements carried out under natural conditions towards standard test conditions (STC), in order to allow a coherent comparison of performances. Two main families of methods have emerged, namely standardized methods, notably the IEC 60891 standard, and alternative methods derived from research, such as those of NREL, ASTM, Blaesser, Castañer, JRC, and Reddy.

The IEC 60891 standard proposes several procedures. The first is empirical and consists of correcting only certain characteristic points (I_sc, V_oc, P_max) from coefficients provided by manufacturers [133]. It is simple to apply and useful when the measurement conditions do not deviate significantly from the STC. However, it is not very precise in cases of highly variable irradiance or extreme conditions because it does not allow complete reconstruction of the curve. The second procedure is based on a one-diode model and allows for the complete translation of the I–V curve [41]. It is more rigorous and suitable for wider variations in temperature and irradiance, but requires a thorough knowledge of the electrical parameters of the module, making its implementation more complex. The third procedure proposes a linear interpolation between two experimental curves measured under different conditions [134]. This is useful when several real curves are available, but it only allows reliable interpolation without possible extrapolation. It is not suitable for estimation outside the measured intervals. The fourth and more recent procedure combines the accuracy of a physical model with improved robustness to correct the entire I–V curve under varied conditions [135]. It is particularly useful for modules with non-linear behavior or in the case of a mismatch, but it requires detailed modeling and a large amount of data. Alternative methods provide more flexible and specific solutions. The NREL method offers simple formulas to correct I_sc, V_oc, and P_max. Widely used in software such as PVsyst or SAM Version 2023.12.17, it is easy to integrate, but it does not allow the generation of the entire I–V curve and relies on linear assumptions that can be limited [140]. The ASTM method E-1036, which is more widespread in the United States, also provides correction formulas, with particular attention paid to the no-load voltage [141]. Although it is reliable for laboratory measurements, it is less commonly used internationally and is poorly suited for field monitoring. The method of Blaesser [142], which is analytical and fast, allows a punctual correction without complete modeling, but it remains imprecise and is very sensitive to measurement errors. This is an interesting method in contexts with low resources. Castañer’s method is mainly interested in the correction of voltage as a function of temperature by introducing the notion of thermal tension [143]. Although it is relevant to understand the thermal influence, it does not treat the entire curve, which limits its use. The JRC and Reddy approaches fall into another category. The JRC method makes it possible to reconstruct the entire I–V curve point by point with great precision. It is well-suited for processing complex data, particularly in monitoring systems or reliability analyses. However, it is greedy for computational resources. Reddy proposed a simplified variant of the JRC method [144] with a reduction in mathematical complexity, making it more suitable for embedded systems. It is useful in environments with limited processing capacity but is less precise than the full JRC approach. In summary, each method has its own advantages and limitations. The procedures of the IEC standard are rigorous and recognized, but vary in complexity and precision. Alternative methods often allow for a more targeted or lighter application, depending on the constraints of the system or analysis context. Thus, the choice of method will depend on the compromise between the desired precision, availability of the necessary parameters, complexity of implementation, and purpose of the analysis (comparison, monitoring, certification, or modeling). A comparative synthesis of these techniques is summarized in

Table 7. Comparison of different I–V curve translation techniques.

Method	Accuracy	Simplicity	Complete Translation	Adapted to Defects	Typical Use
IEC Procedure 1	low	better	No	Low	Simple Comparison
IEC Procedure 2	better	no	Yes	Average	Performance Studies
IEC Procedure 3	Average	Average	Partial	Low	Experimental Interpolation
IEC Procedure 4	better	no	Yes	Good	In-Depth Diagnostics
NREL	Average	Average	No	Low	PVsyst Software v7.3, 2024
ASTM	Average	no	No	Low	Experimental US Analysis
Blaeser	low	better	No	Low	Rapid Correction
Castaner	low	Average	No (Voc only)	Average	Thermal Analysis
JRC	better	no	Yes	Good	High-Resolution Monitoring
Reddy	Average	Average	Yes	Average	Embedded Systems

.

4. Conclusions and Future Perspectives

Photovoltaic (PV) technologies play an increasingly crucial role in global energy production. Therefore, it is essential to know the performance of these technologies under real operating conditions compared with the standard data provided by manufacturers. This article provides an overview of the different electrical I–V characterization techniques for PV modules, allowing verification of their guarantees and energy performance.

Both indoor and outdoor characterizations are used to ensure the performance of photovoltaic installations. Indoor characterization is the method used by manufacturers to determine the characteristics of PV modules, as indicated on nameplates and data sheets, under standard test conditions (STC) and nominal irradiation and temperature conditions (NOCT). These tests also define the temperature and irradiation coefficients, as well as corrections to the performance parameters of PV modules. However, because the spectral distribution of sunlight is not as stable outdoors as in a laboratory, solar simulators are used to ensure the repeatability of the measurements. Several types of solar simulators exist, such as integrated thermal chamber simulators, simulators equipped with irradiation sources, and artificial cooling and collecting surfaces for the assembly of modules.

In order to take into account the environmental and climatic parameters, and given the high cost of solar simulators, various characterization techniques are used under real-world conditions. Each method has its own specific criteria, which must be adapted to the external conditions in order to guarantee a reliable performance study of PV modules. These criteria are mainly related to the reliability of the results, both from the point of view of load disturbances and environmental influences, as recorded by I–V curve tracers. Other factors, such as simplicity, cost, and sustainability of the system, are also taken into account. It should be noted that some methods are suitable for low-power PV generators, others for high-power, and some are suitable for all types of photovoltaic fields, regardless of their size.

In addition to I–V curve studies, it is also possible to characterize a PV module by observing its output drop in power or energy, depending on the levels of degradation or impacts of environmental parameters such as temperature, irradiation, and dust accumulation. These observations provide a better understanding of how these factors influence the performance of the module over time.

It is important to note that a PV module performance study can only be considered reliable if the characteristic parameters measured under natural conditions are compared with those provided by the manufacturer. To perform this comparison, the parameters measured under real conditions must be translated into standard test conditions (STC). This is done using translation or extrapolation equations, such as those defined in the international standards for characterization of PV modules.

This article provides useful information for technicians, engineers, and researchers on the different techniques, methods, and protocols of characterization in outdoor conditions to study and evaluate the impact of external parameters on the performance of photovoltaic modules.