Analysis of the Weather Impact on the Photovoltaic Panel Efficiency Based on Machine Learning Techniques

The efficiency and power output of solar panels are critically influenced by weather conditions, making accurate predictions essential for optimizing photovoltaic power plants. This paper presents a comprehensive analysis of solar panel performance based on real weather data using machine learning techniques. It focuses on analyzing the impact of climatic conditions on the efficiency of photovoltaic panels at the power plant in Tozeur, Tunisia. Historical weather data, including air temperature, humidity, solar irradiance, wind speed, and surface pressure, were used as input variables in various machine learning models, such as ANN, FFNN, and Random Forests, to predict the efficiency of solar panels under varying weather conditions. The results indicate that certain weather variables, particularly solar irradiance and air temperature, significantly impact the efficiency of solar panels. The findings suggest that machine learning can effectively predict solar panel efficiency, providing valuable insights into the most important meteorolog-ical variables affecting panel performance and indicating their potential for improving photovoltaic power plant efficiency.