Solar Energy forecast models based on gradient boosting algorithms
Veeraraghava Raju Hasti (1)
(1) School of Mechanical Engineering, Purdue University, West Lafayette, IN USA 47907
This paper presents machine learning-based forecasting models for predicting the power output from solar photovoltaic cells based on the given weather data. The weather data was obtained from the National Solar Radiation Data Base and the solar energy production data was obtained from the University of Massachusetts Amherst’s (UMass Amherst) energy dashboard. The CatBoost and XGBoost models are trained and validated in this study. The performance of these two models is evaluated. The R2 value for CatBoost is 0.89 and XGBoost is 0.86. The results of this study show that CatBoost and XGBoost can be used to predict the power output from solar panels.