Weaknesses and Improvements of the Extended Kalman Filter for Battery State-of-Charge and State-of-Health Estimation

Battery management systems (BMS) are essential for ensuring battery performance and safety. Accurate estimation of the State of Charge (SOC) and State of Health (SOH), for example, are critical. However, utilizing the conventional Extended Kalman Filter (EKF) for SOC and SOH co-estimation is often challenging due to problems such as overconfident covariance estimation, overly simplistic assumptions about pro-cess noise and measurement noise covariance matrices, and the shift of the open circuit voltage (OCV) curve as the cell ages. To address these issues, this paper introduces an improved EKF design for co-estimating the SOC and SOH. The proposed approach incorporates innovative strategies to counteract covariance pitfalls, calculates the optimal covariance matrix configuration objectively, and incorporates OCV shifts from aging. Comparative simulations underscore the superiority of our method against traditional EKF and Unscented Kalman Filter (UKF) techniques. The code is available at https://github.com/Shida-Jiang/EKF_UKF_SOCSOH_estimation.