On-board State-of-health Estimation Based on Charging Current Analysis for LiFePO4 batteries

This paper proposes an on-board state-of-health (SoH) estimation method based on the charging current curve during the constant-voltage (CV) stage. Through analyzing the battery aging cycle data, the equivalent time constant based on the equivalent circuit model (ECM) is introduced to describe the dynamic characteristic of the current curve during the CV stage. In addition, only partial current curves are adopted to identify the equivalent time constant. Four lithium iron phosphate (LiFePO4) batteries are employed under test to validate the feasibility of the developed method. The results show that regardless of the adopted data length, the linear regression identified from one battery can be used to indicate the SoH of other three batteries within 1.5% error bound except a few outliers.