Harnessing Machine Learning with Advanced Linear Regression Models to Forecast PV System

This study investigates how environmental factors impact photovoltaic systems’ outputs, through a one-year data collection. Nineteen machine learning linear regression models were employed to predict power, current, and voltage outputs, including Ridge Cross Validation, Lasso, ElasticNet, Orthogonal Matching Pursuit, Bayesian Regression, ARD Regression, SGD, PA, RAC, TheilSen, and Huber. The models' performances were evaluated and compared. The findings emphasize the potential of advanced machine learning techniques in optimizing PV system performance under varied environmental conditions.