A Numerical Study of Lithium-ion Battery Degradation Feature Extraction based on Transfer Function

Abstract

Lithium-ion battery has been widely applied in various industrial fields including electrical vehicles (EVs), portable electronics and spacecraft. As the key part of the power supply system, the health state of the battery directly influence the reliability and safety of the host system. The degradation feature extraction approach is the vital part for battery state of health (SOH) estimation. The lithium-ion battery can be equalized as an electric circuit consists of a current source and several capacitances and resistances. These parameters varies based on the battery state of charge (SOC) and SOH. In other words, the battery performance degradation leads to the variation of the parameters in battery equivalent circuit model (ECM) which will directly influence the battery transfer function. Thus, this paper quantitative analysis of degradation characterization capability of parameters in battery equations. For the first-order ECM, the transfer function is established and the parameters are identified based on the battery hybrid pulse power characteristic test. Three different kinds of correlation analysis methods are applied in to evaluate the effectiveness of the parameters in transfer function for degradation representation.