Performance Prediction and Estimation of Battery State of Charge Using DNN for Electric Vehicles

Electric vehicles (EVs) have seen a rapid rise in popularity due to their environmental benefits and energy efficiency. However, accurate prediction and estimation of the state of charge (SOC) of batteries used in EVs is critical for efficient battery management. According to a report by Research and Markets, the global market for EV batteries is expected to reach USD 84.5 billion by 2025, indicating the importance of developing accurate SOC prediction models. This study focuses on the development of deep neural network (DNN) models, namely long short-term memory (LSTM) and convolutional neural network (CNN), for the purpose of predicting and estimating the state of charge (SOC) of batteries used in electric vehicles (EVs). The dataset used for training and testing the DNN models comprises of 8 lakh data points and contains five features, namely voltage, current, temperature, average voltage, and average current, with the target variable being the SOC. The performance of the models is evaluated using various metrics, such as mean absolute error (MAE), maximum absolute error, and root mean squared error (RMSE). The study demonstrates the potential of using DNN models for improving the performance and efficiency of battery management systems in EVs.