Machine learning based models for solar energy

Abstract

Photovoltaic (PV) technology is one of the most promising forms of renewable energy. However, power generation from PV technologies is highly dependent on variable weather conditions, which are neither constant nor controllable, which can affect grid stability. Accurate forecasting of PV power production is essential to ensure reliable operation within the power system. The primary challenge of this study is to accurately predict photovoltaic energy production, considering that weather conditions, such as irradiance, temperature, and wind speed, are random variables. The key contribution of this article is developing a machine learning model to predict the energy production of a real PV power plant in Algeria. Using real measurements sourced from the Center of Renewable Energy Development (CDER) in Adrar, Algeria, in 2021. The data are from two PV power plants located in harsh desert climate conditions. The results presented in this study offer a comparison of several predictive methods applied to real-world data from a PV power plant situated in the Saharan Region. Our findings reveal that the artificial neural network (ANN) model yields the most accurate predictions of 94.96%, with the smallest prediction error: root mean square (RMSE) and mean absolute error (MAE) are 7.78% and 3.80%, respectively.