SOE Estimation of Lithium-Ion Battery Based on Strong Tracking Extended Kalman Filter

The state of energy (SOE) serves as an essential parameter in power batteries for new energy vehicles, significantly influencing the energy management strategy of battery management systems (BMS). This study presents a model-based strong tracking filter algorithm to enhance the accuracy and adaptability of SOE estimation amidst changing operating conditions. The second-RC networks model is utilized to identify battery parameters via the hybrid power pulse test. Additionally, for scenarios with randomly varying working conditions, an online SOE estimation approach utilizing the strong tracking extended Kalman filter algorithm is introduced, ensuring robust tracking performance, especially under sudden changes in conditions. The algorithm's validity is confirmed through simulation across various working conditions, demonstrating superior accuracy and robustness compared to the conventional Kalman filter algorithm. Overall, the proposed algorithm meets the requirements of SOE estimation in online scenarios, offering enhanced tracking performance and reliability.