A hybrid machine learning approach for real time wind power prediction for micro grid energy management

Abstract In today’s dynamic energy, it is difficult to predict the accurate forecasting of wind energy generation. In order to accurately estimate wind energy, this study investigates the use of a hybrid machine learning system that combines XG-Boost and Linear Regression. By utilizing past data on temperature, humidity, wind speed and wind direction the hybrid model outperforms separate algorithms in terms of predictions. After a thorough analysis with metrics like mean absolute error (MAE), root mean square error (RMSE) and R 2 error, the hybrid algorithm regularly performs better than its competitors in capturing the non-linear and linear interactions present in wind energy systems. The present study advances the methodology for renewable energy forecasting by highlighting the effectiveness of hybrid machine learning approaches in enhancing the accuracy of wind energy predictions. The results highlight the hybrid model’s potential to improve wind energy production’s efficiency and dependability, supporting sustainable and effective utilization of renewable energy resources. We have employed five machine learning algorithms such as Linear Regression, Random Forest, Support Vector Machine, and XG-Boost and hybrid model for wind energy forecasting.