Implementation of Machine Learning in Battery Management System of Electric Vehicles

This research presents a machine-learning approach for predicting the State of Charge (SOC) of batteries used in electric vehicles.SOC prediction plays a key role in the Battery Management System (BMS) as it helps ensure safe operation, proper charging, and improved battery life.In this study, a dataset containing voltage, current, temperature, and SOC is used to develop two models: Linear Regression (LR) and Support Vector Regression (SVR).The data is pre-processed and divided into training and testing sets to evaluate the performance of both techniques.Linear Regression provides a basic linear model for SOC estimation, making it easy to interpret and implement.SVR, on the other hand, is capable of handling more complex relationships between the input values and SOC.After testing, the results show that both models give reliable predictions, with SVR showing a slight advantage when the data contains non-linear patterns.The findings highlight the usefulness of machinelearning methods in building intelligent BMS systems and contribute to improving SOC estimation in electric vehicles.