Adaptive robust unscented Kalman filter with recursive least square for state of charge estimation of batteries

The state of charge (SOC) estimation is the core of any battery management system (BMS). However, the accurate SOC estimation is always a challenging task because it cannot be measured directly with sensors. An adaptive robust unscented Kalman filter (ARUKF) with recursive least square (RLS) is proposed to improve the robustness and accuracy of SOC estimation in this paper. In the proposed method, RLS is used for identification of the parameters of battery. Then, based on the identified parameters, ARUKF is designed to estimate the SOC of battery. The ARUKF is developed using embedding the unscented transformation (UT) technique and H∞ filtering into the unscented Kalman filter (UKF). The knowledge of noise distributions does not require in the proposed method, and the noises can be non-Gaussian, so it has less limitation in actual applications. In the proposed method, to improve more performance, the covariance of measurement and process noise are tuned. The process and measurement noise covariance can be adaptively tuned that improves the stability and accuracy of filter. The proposed method is evaluated under different real-time conditions. The results of proposed method are compared with those of extended Kalman filter (EKF) and UKF. The results show that the proposed method can achieve better SOC estimation accuracy, especially when the noise statistics are unknown and non-Gaussian.