State-of-Charge Co-estimation of Li-ion Battery based on on-line Adaptive Extended Kalman Filter Carrier Tracking Algorithm

Abstract

Li-ion batteries as a source of energy in electric vehicles (EV) and hybrid electric vehicles (HEV) are receiving more attention with the worldwide demand for energy conservation and environmental protection. In this paper, an improved State-of-Charge (SOC) co-estimation algorithm based on the second-order RC equivalent circuit model is proposed. Firstly, Forgetting Factor Recursive Least Squares (FFRLS) algorithm is adopted to realize on-line parameter identification of the model. Secondly, SOC is estimated with identified parameters by adaptive extended Kalman filter carrier tracking (AEKF) algorithm based on innovations and residuals. The results of two discharge experiments in different conditions show that the co-estimation algorithm has a higher estimation accuracy, convergence speed and robustness compared with off-line AEKF SOC estimation algorithm, which is more suitable for on-line estimation of electric vehicle SOC.