Predicting Power Generation from Photovoltaic Energy System
摘要
Solar photovoltaic energy production prediction is a constantly evolving field of research that uses a variety of methods and approaches to maximize the efficiency of PV systems and ensure the stability of the power grid. This work shows a predictive study of photovoltaic solar energy production by training various regressors such as Lasso, Ridge and ElasticNet and a custom Multilayer Perceptron. The best performance with the test set was achieved by the custom Multilayer Perceptron with a R2 of 0.782. These results demonstrate the effectiveness of advanced machine learning techniques in improving solar energy forecasting.