Comparative Analysis of Statistical and Machine Learning Models for Solar Energy Production Forecasting and Model Selection.

Abstract

Efficient energy forecasting methods have become the focal point to ensure sustainable energy production. The present study aimed to use cutting-edge machine learning models along with blended statistical models to precisely predict solar energy production. Three data-based Machine learning models-Random Forest, Support vector machine, and XGBoost-and three statistical models- Linear regression, ARMA, and ARIMA models were employed. Statistical indices such as Mean absolute error (MAE), Mean square error (MSE), and Root mean square error (RMSE) were used to determine the most precise predictive model. The results demonstrated that XGBoost was the most precise predictive model, with MAE value of 0.1618 MSE of 0.1542, and an RMSE of 0. 3927. It is recommended that the use of blended Machine Learning and statistical forecast models would be a valuable tool for policymakers, solar energy researchers, and solar farm developers.