Sensitivity analysis and identification of battery physicochemical model parameters under different temperature impedances

Lithium–ion batteries are widely used vehicle energy storage batteries globally, and their reaction mechanism directly influences the safety and performance of energy storage systems. Simulating the actual internal state of the battery through a simulation model has become a crucial approach. Despite the high level of physical interpretation provided by the physicochemical models of the battery, the parameterization process poses significant challenges. Impedance spectroscopy at various temperatures is analyzed in this study to assess the sensitivity of battery model parameters. The goal is to investigate the changes in model parameter sensitivity at different temperatures and identify model parameters under various states of charge. Conducting a sensitivity analysis of model parameters allows for the selection of highly sensitive and appropriate parameters, which enhances the accuracy of the model output. In addition, the gray wolf optimization algorithm is utilized to identify the parameters of the battery model. The identified parameter set is then applied to the battery P2D model to simulate the impedance data. The root mean square error between the simulation and experimental data of the model is less than 0.241 mΩ, and the average absolute error is less than 0.185 mΩ. This method can accurately identify model parameters, which results in high model simulation accuracy. Findings from the parameter sensitivity analysis provide valuable references for parameter identification and state estimation considering temperature effects.