State of Health Estimation of Lithium Batteries Based on Electrochemical Impedance Spectroscopy and Data-Driven Approaches

This study proposes a method for estimating the state of health (SOH) of lithium batteries based on Electrochemical impedance spectroscopy (EIS) and Data-driven techniques. First, an equivalent circuit model was developed based on EIS. The least squares method was then applied to identify the equivalent circuit parameters, which were extracted as health features. Next, a long short-term memory (LSTM) neural network was established, and the sparrow search algorithm (SSA) was used for network parameter optimization, enabling accurate SOH estimation. The results indicate that the root mean square error (RMSE) of the SOH prediction, using features extracted from the EIS equivalent circuit, is less than 0.0088. After SSA optimization, the RMSE of the LSTM model was reduced by 35.4% compared to its baseline model.