Renewable power capacity sets records annually, driven by solar photovoltaic power, which accounts for more than half of all renewable power expansion in 2021. In this sense, photovoltaic system design must be correctly defined before system installation to generate the maximum quantity of energy at the lowest possible cost. The proposed study analyses the oversizing of the solar array vs. the capacity of the solar inverter, seeking low clipping losses in the inverter. A real 4.2 kWp residential PV installation was modelled and validated using the software SAM and input data from different sources, such as a weather station for weather conditions, ESIOS for electricity rates, and FusionSolar to obtain energy data from the PV installation. Once data were validated through SAM, the DC to AC ratio was varied between 0.9 and 2.1. The azimuth and slope sensitivity analyses were performed regarding clipping inverter losses. Results have been evaluated through the energy generated and the discounted payback period, showing that, depending on the weather conditions, slope, and azimuth, among others, it is advisable to increase the DC to AC ratio to values between 1.63 and 1.87, implying low discounted payback periods of about 8 to 9 years. In addition, it was observed that inverter clipping losses significantly vary depending on the defined azimuth and slope.

Accurate temperature prediction in bifacial photovoltaic (PV) modules is critical for optimizing solar energy systems. Conventional models face challenges to balance accuracy, interpretability, and computational efficiency. This study addresses these limitations by introducing a symbolic regression (SR) framework based on genetic algorithms to model nonlinear relationships between environmental variables and module temperature without predefined structures. High-resolution data, including solar radiation, ambient temperature, wind speed, and PV module temperature, were collected at 5 min intervals over a year from a 19.9 MW bifacial PV plant with trackers in San Marcos, Colombia. The SR model performance was compared with multiple linear regression, normal operating cell temperature (NOCT), and empirical regression models. The SR model outperformed others by achieving a root mean squared error (RMSE) of 4.05 °C, coefficient of determination (R²) of 0.91, Spearman’s rank correlation coefficient of 0.95, and mean absolute error (MAE) of 2.25 °C. Its hybrid structure combines linear ambient temperature dependencies with nonlinear trigonometric terms capturing solar radiation dynamics. The SR model effectively balances accuracy and interpretability, providing information for modeling bifacial PV systems.

El presente trabajo desarrolló el análisis experimental comparativo de los datos reales de producción de energía de una planta solar fotovoltaica de 72KWn, con simulaciones hechas en PVsyst, PVGIS y SAM. El periodo de medición fue dividido entre marzo de 2021 a febrero de 2022 y de marzo de 2022 a febrero de 2023. Se encontró que la simulación desarrollada en SAM presentó el error cuadrático medio más bajo para todo el periodo de medición en comparación con PVsyst y PVGIS, presentando valores respectivos de para SAM, para PVGIS y para PVsyst. Además, se concluyó que un aumento en la temperatura ambiente puede influir en la producción de un sistema fotovoltaico, asi también el uso de bases de datos ambientales actualizadas, puede resultar en cálculos más precisos de la producción proyectada.

The thermal behavior of the single-axis tracked bifacial photovoltaic (PV) module is important for efficient energy extraction in large-scale power plants, especially in tropical regions under high irradiation and high ambient temperature. However, it is difficult to accurately predict their operating temperature due to the complex interaction between environmental variables and the characteristics of solar tracking. The available models, ranging from empirical correlations and computational fluid dynamics (CFD) simulations to machine learning methods, face challenges in terms of accuracy, interpretability, and computational load. This gap is addressed in this study, with the development of a modeling methodology based on symbolic regression (SR) utilizing genetic algorithms (GA) towards obtaining an explicit, interpretable Equation for the prediction of the PV module temperature in single-axis tracking systems. One year of data was collected at 5-minute intervals from a 19.9 MW PV plant located in San Marcos, Colombia, consisting of measurements for solar radiation, ambient temperature, wind speed, and module temperature. The constructed SR GA model achieved satisfactory prediction accuracy compared to classic models with the best root mean square error (RMSE = 4.14 °C) and R² (0.91) on the test data set. These results compare favorably with results from MLR (RMSE = 4.31 °C, R² = 0.90), the standard industry NOCT model (RMSE = 8.59 °C, R² = 0.60), and the empirical Skoplaki I model (RMSE = 5.92 °C, R² = 0.81). The resulting symbolic equation directly characterizes the effects of nonlinear solar radiation, ambient temperature, and wind speed, providing greater physical insight into the thermal dynamics of the system. An important finding is that the maximum temperature of the bifacial module is reached around 14:00h, probably due to the accumulation of temperature caused by solar tracking, which contrasts with what occurs in fixed-tilt monofacial technology. This study demonstrates that the symbolic regression technique with a genetic algorithm kernel can produce accurate, interpretable, and computationally economical models for advanced photovoltaic systems.

Accurate evaluation of energy production in photovoltaic (PV) systems is critical for renewable projects, especially in tropical climates where environmental factors such as temperature significantly affect performance. Although commercial simulation tools exist (photovoltaic geographic information system (PVGIS), PVsyst, and system advisor model (SAM)), previous studies have identified notable deviations between their predictions and actual data, particularly in tropical climates. Moreover, these investigations are usually limited to short periods (one year) and do not systematically compare multiple tools under interannual conditions. This study evaluates the accuracy of PVGIS, PVsyst, and SAM in predicting the energy production of a PV installation in a tropical equatorial climate for 24 months to identify the most suitable tool for this context. Monthly energy production data were collected from a PV plant in Monteria, Colombia, equipped with 240 modules and two 36 kW inverters. Simulations were performed using the most recent PVGIS, PVsyst, and SAM versions. Accuracy was evaluated using metrics such as root mean square error (RMSE) and mean absolute error (MAE). SAM showed the highest accuracy, with an overall RMSE of 1,993.71 kWh and MAE of 1,615.87 kWh, followed by PVGIS (RMSE: 2,076.65 kWh, MAE: 1,830.84 kWh) and PVsyst (RMSE: 3,546.18 kWh, MAE: 3,250.17 kWh). The results highlight that SAM provides estimates closer to the real data and less dispersion than other tools. This study contributes to the renewable energy field by systematically comparing simulation tools in an understudied tropical context. The findings emphasize the importance of selecting appropriate software according to the specific environmental conditions of the project, thus optimizing the design and efficiency of PV systems in tropical regions.