EV Charging Optimization through Wind Power Forecasting using Machine Learning Model

Abstract

Accurate prediction of wind  power is essential  for  effective planning  and  management  of electric vehicle (EV) charging, especially as renewable energy sources become more widely used. This study explores a machine learning approach using a Quantile Regressor algorithm to forecast wind power and determine optimal EV charging times. Different optimization techniques are tested within the algorithm to improve its prediction accuracy. The performance of each method is evaluated using a wind power dataset, with the goal of minimizing prediction errors, measured by Mean Squared Error (MSE). The results highlight the impact of optimizer choice on model performance and identify the most effective configuration for accurate wind power forecasting. This approach can help align EV charging schedules with renewable energy availability, supporting more efficient and sustainable energy use.