Exploring the Potential of Machine Learning for EV Energy Demand Forecasting in Bangladesh

The expanding use of electric vehicles (EVs) in Bangladesh poses a substantial challenge to the current power grid. This study which intends to address this issue, creates a machine learning-based model that reliably forecasts EV energy use at charging stations. Using a dataset from California City as a proxy, we investigate the performance of different machine learning techniques, such as Random Forest (RF), Decision Tree (DT), and Gradient Boosting Regression (GBR). Gradient Boosting Regression outperformed other methods, with a R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.999959 and a mean absolute error (MAE) of 0.0071. The ramifications of this research go beyond making correct forecasts. This study helps to ensure the successful integration of EVs into Bangladesh’s transportation sector by optimizing energy management, assessing grid impacts, and supporting sustainable infrastructure development. This paper is a pioneering effort to anticipate EV energy use in emerging countries, giving valuable data for policymakers and industry professionals.