A COMPARATIVE ASSESSMENT OF MACHINE LEARNING MODELS FOR PREDICTING WIND SPEED

Abstract

Renewable energy is one of the most critical issues of continuously increasing electricity consumption which is becoming a desirable alternative to traditional methods of electricity generation such as coal or fossil fuels. This study aimed to develop, evaluate, and compare the performance of Linear multiple regression (MLR), support vector regression (SVR), Bagging and random forest (R.F.), and decision tree (CART) models in predicting wind speed in Southeastern Iran. The data used in this research is related to the statistics of 10 minutes of wind speed in 10-meter, 30-meter, and 40-meter wind turbines, the standard deviation of wind speed, air temperature, humidity, and amount of the Sun's radiation. The bagging and random forest model with an RMSE error of 0.0086 perform better than others in this dataset, while the MLR model with an RMSE error of 0.0407 has the worst.