An Efficient Machine Learning Based Optimization Framework to Analyse the Short-Term Solar Energy

A short-term prediction of solar energy is absolutely necessary in order to maximize the efficiency with which solar power plants utilize energy. The aim of this research is to enhance the precision and efficiency of solar energy forecasts by introducing an innovative system that merges the Random Forest Classifier (RFC), a machine learning method, with the Election-Based Optimization Algorithm (EBOA). The proposed system utilizes the Request for Comments process for feature selection and model creation. Additionally, the Evolutionary Biogeography-Based Optimization Algorithm (EBOA) is employed to fine-tune the model parameters, leading to significantly improved prediction performance. Typical measures such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and R-squared (R2) coefficient are used to evaluate the effectiveness of the proposed model. Test results show that the RFC-EBOA method performs better, than existing approaches in terms of predicting efficiency. The results imply that the proposed framework excels over methods, in prediction precision.