Publication

Abstract :

Accurate wind speed prediction at different heights is critical for applications such as wind energy generation and meteorology. This study investigates the simultaneous prediction of wind speed at 10m and 100m using support vector machine, linear regression, gradient boosting, and random forest models. A dataset comprising historical wind speed measurements and relevant meteorological parameters is utilized for training and evaluation. Model performance is assessed using standard metrics such as mean square error and R2 score. Among the models tested, random forest demonstrates superior predictive accuracy, effectively capturing the non-linear dependencies in wind speed variations. These findings highlight the effectiveness of ensemble-based approaches for wind speed prediction and provide insights for optimizing wind energy deployment.