Fast Static Electrochemical Impedance Spectra Prediction from the Dynamic Impedance of Lithium-Ion Batteries

Lithium-ion batteries (LIBs) are widely used in modern energy storage systems due to their high energy density and long cycle life. Electrochemical Impedance Spectroscopy (EIS) is a powerful non-destructive method for assessing battery performance and aging, and has traditionally relied on static EIS measurements performed under steady-state conditions. However, obtaining steady-state data is time-consuming and can interfere with battery operation. Dynamic electrochemical impedance spectroscopy (DEIS) addresses this issue to some extent, however, interpreting the transient nature of these data is challenging. In this study, we propose a new method to predict static EIS from DEIS using a long short-term memory neural network, thus saving measurement time. The results show that the complex mapping between dynamic and static EIS can be accurately predicted from static EIS by a deep learning approach with Mean Absolute Error of 0.0193 and 0.0080 for its real and imaginary parts, respectively. This study makes significant progress in simplifying the EIS acquisition, which makes battery diagnosis more efficient and flexible.