Battery State-of-Charge Estimation Based on Fuzzy Dual Kalman Filter Algorithm

Abstract

Extended Kalman filter (EKF) algorithm can be used as a model-based approach to estimate the state of charge (SOC) of batteries, and the accuracy of SOC estimation can be affected by the error of battery model. Moreover, a large current or current change leads to a large model error. A fuzzy dual Kalman filter (Fuzzy-DKF) algorithm was proposed to estimate SOC in this paper. Specifically, Thevenin equivalent circuit model was established, and was transformed to autoregressive exogenous (ARX) model, the parameters of ARX model were identified by Kalman filter (KF) algorithm; current and current change were monitored in real time and a fuzzy control system was built to reduce the impact of model error by modifying Kalman gain; EKF algorithm was adopted to estimate SOC. The results indicate that the accuracy of SOC estimation based on Fuzzy-DKF algorithm is better than the accuracy of SOC estimation based on EKF algorithm or dual Kalman filter (DKF) algorithm.