Improved Diagnosis of Lithium-Ion Battery Health in Electric Vehicles via a Hybrid Deep Learning Model Incorporating Wavelet Transform and Attention Mechanism

Abstract

As the adoption of lithium-ion batteries (LIBs) in electric vehicles increases, ensuring their reliability and safety is essential. The Battery Management System (BMS) is vital for accurately evaluating the State of Health (SOH) of these batteries to ensure safe vehicle operation. To address this, a novel time series model for predicting SOH in Li-ion batteries is proposed. This model combines wavelet transform with a hybrid architecture of convolutional neural networks (CNNs) and long short-term memory (LSTM) networks, enhanced by an attention mechanism to improve performance. Experimental results indicate that the wavelet-enhanced LSTM method significantly improves prediction accuracy. This study presents a promising approach to enhance the reliability and efficiency of electric vehicle battery systems, supporting broader adoption and sustainability in electric transportation.