The Joint Estimation of SOC-SOH for Lithium-Ion Batteries Based on BiLSTM-SA
Lingling Wu, Chao Chen, Zhenhua Li, Zhuo Chen, Hao Li
Abstract
Lithium-ion batteries are commonly employed in energy storage because of their extended service life and high energy density. This trend has coincided with the rapid growth of renewable energy and electric automobiles. However, as usage cycles increase, their effectiveness diminishes over time, which can undermine both the system’s performance and security. Therefore, monitoring the state of charge (SOC) and state of health (SOH) of batteries in real time is particularly important. Traditional SOC calculation methods typically treat SOC and SOH as independent variables, overlooking the coupling between them. To tackle this issue, the paper introduces a joint SOC-SOH estimation approach (BiLSTM-SA) that leverages a bidirectional long short-term memory (BiLSTM) network combined with a self-attention (SA) mechanism. The proposed approach is validated using a publicly available dataset. With the SOH taken into account, the MAE and RMSE of the SOC are 0.84% and 1.20%, showing notable increases in accuracy relative to conventional methods. Additionally, it demonstrates strong robustness and generalization across datasets with multiple temperatures.