Parameter Sensitivity Analysis for A Thermal Coupling Simplified Electrochemical Model and Health Feature Extraction

Battery state of health is critical for a battery management system because it can indicate how much energy can be effectively extracted. How to estimate SOH accurately becomes a big issue. This work develops a thermal coupling simplified electrochemical model that has good parameter identifiability. A sensitivity analysis is conducted to determine which parameters make more contributions on model simulation. The variations of battery terminal voltage and surface temperature caused by the changes of the parameters are used to assess their sensitivities. Then, an aging test is conducted to obtain the battery with different SOHs. Lastly, three sensitive parameters are selected as internal health features according to their correlation coefficients with SOH, which provides more health information for further SOH estimation.