Partial Charging-Based Health Feature Extraction and State of Health Estimation of Lithium-Ion Batteries

Abstract

The state-of-health (SOH) is a critical indicator in the battery management system of lithium-ion batteries (LIBs). Charging feature-based SOH estimators have attracted much attention due to the stability and controllability of working conditions. However, the uncertainty raised from partial charging substantially rolls up the difficulties in feature extraction. A new incremental capacity (IC) feature-based SOH estimation method using the partial constant current (CC) charging data of LIB is proposed in this article. First, the linear model between the specific features of interest (FOI) and the SOH is calibrated by fitting the revised Lorentzian function-based voltage-capacity (RL-VC) model with the full-range CC charging data. Second, the corresponding FOI is extracted from the IC curves by fitting the RL-VC model with the partial CC data. Then, the SOH can be directly interpolated based on the calibrated FOI-SOH linear model. The estimated SOH accuracy holds within 5% when different partial CC datasets are used, which indicates the proposed SOH estimator can be reliably implemented in practical applications.