A Joint State Estimation Framework for Lithium-ion Batteries based on Hybrid Method

Abstract

Accurate estimation of state of charge (SOC), state of health (SOH) and remaining useful life (RUL) of Lithium-ion batteries is of great importance. This paper proposes a estimation framework based on charging voltage segment and hybrid method combining with equivalent circuit model and data-driven algorithms to achieve the accurate estimation of SOC, SOH and RUL of the long life cycle of the battery. The charging voltage segment, with the merit of flexibility and available, which contains much information for battery degradation, is suitable for battery modeling and state estimation. The rising time of the optimal charging voltage segment which has the highest correlation with the present capacity is extracted as health feature (HF) to estimate the SOH with least squares support vector machine (LSSVM), and the identified resistance and capacitance parameters acquired by fitting the charging voltage segment with equivalent circuit model and estimated capacity acquired by LSSCM are used to estimate SOC with the unscented Kalman filter (UKF) algorithm in the discharge stage for this cycle; Gaussian process regression (GPR) is used to prediction the trend of HF, which are combined with established LSSVM model to estimate RUL. Experimental results show that the proposed method can estimate SOC, SOH and RUL jointly with high accuracy.