An iteratively reweighted robust regression model for PV electrical and thermal performance using actual measurement data

The performance of photovoltaic (PV) modules is inherently dynamic, governed by a complex interplay of meteorological factors such as solar irradiance, temperature, and atmospheric conditions. Accurately and effectively predicting this performance remains a crucial area of research. This study aims to establish the correlation between meteorological parameters and power output as well as cell temperature. For this purpose, a prediction model was developed using iteratively reweighted robust regression to estimate both power output and cell temperature. Robust regression models were developed using real measurement data from the PV system and weather station established at the National Institute of Standards and Technology (NIST). The coefficient of determination (R 2 ) and root mean square error (RMSE) of the proposed model for power output are 0.9870 and 0.0542, respectively. For cell temperature, these values are 0.8443 and 0.2194, respectively. The R 2 and RMSE statistics of the proposed models indicate successful performance for the climatic conditions studied. Additionally, a series of experiments were conducted to evaluate the sensitivity of the proposed models to the variation in PV performance across regions. Therefore, the developed model is recommended as a powerful tool for predicting the electrical and thermal performance of PV systems.