Online state-of-charge and capacity co-estimation for lithium-ion batteries under aging and varying temperatures

Accurate estimation of the state of charge (SOC) and capacity under various aging and temperature conditions is crucial for an effective battery management system. However, numerous studies on the co-estimation of SOC and capacity have assumed that capacity remains constant within a single cycle, often overlooking time-varying temperature conditions which limits their applicability to real-world scenarios. To address this limitation, a novel online SOC and capacity co-estimation method was developed that not only functions effectively under aging conditions but also adapts to time-varying temperature conditions by directly capturing trends in the changing capacity. The proposed method identifies the equivalent circuit model parameters using the forgetting factor recursive least square algorithm. The terminal voltage and internal resistance are then integrated into the extended Kalman filter for accurate estimation. The efficiency of the method was validated through experimental investigations under aging and time-varying temperature conditions. In the aging experiments, the method achieved SOC and capacity estimation errors below 0.75% and 3.15%, respectively. In the time-varying temperature experiments, SOC and capacity estimation errors were below 0.58% and 1.17%, respectively. Compared to traditional methods, the proposed algorithm achieved superior performance under time-varying temperature conditions, ensuring reliability in real-world conditions.