Advanced Machine Learning Algorithms for Battery State-of-Charge Estimation in Electric Vehicles

The rapid adoption of electric vehicles (EVs) as a cleaner alternative to gasoline- and diesel-powered transportation has introduced significant challenges, particularly in the accurate estimation of battery state-of-charge (SOC). In recent years, the application of machine learning (ML) techniques to battery SOC estimation has gained increasing attention; however, there remains a lack of understanding regarding the performance of different techniques under varying conditions. This study focuses on evaluating and comparing various ML techniques, including deep learning (DL) algorithms, for battery SOC estimation. The goal is to identify the most accurate SOC estimation method to enhance the efficiency and sustainability of EV battery systems. The algorithms examined in this study include four widely used techniques: Random Forest (RF), Long Short-Term Memory (LSTM), Feedforward Neural Network (FFNN), and Convolutional Neural Network (CNN). These algorithms are trained, tested, and validated on publicly available battery datasets, with their performance evaluated using metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), R-squared (R²), and Root Mean Squared Error (RMSE). Our results demonstrate that the FFNN method outperformed the other algorithms in terms of prediction accuracy and overall performance.