Simultaneous estimation of SOC and capacity for lithium-ion battery packs throughout full lifespan based on migration modeling

Accurate estimation of state of charge (SOC) and capacity is critical for ensuring safe and reliable operations of lithium-ion battery packs, especially when subjected to wide temperature ranges and capacity degradation. To address these influences, this study develops a migration model combined with the risk-minimizing particle filter (RMPF) algorithm for individual cell SOC and capacity estimation, and then applies the Vmin + Vmax model (VVM) for the pack state estimation. Firstly, a migration framework based on a first-order resistance-capacitance (RC) equivalent circuit model is established to enable automatic updating of model parameters, accurately capturing the effects of temperature variations and battery aging on battery state estimation. Secondly, the capacity parameterization is conducted to address the rapid SOC change and the slow capacity variation and achieve mutual decoupling in simultaneous estimation of SOC and capacity for individual cells, which is then estimated by the RMPF algorithm. Finally, a VVM with incorporated weighting factors is employed to compensate cell inconsistency. To assess the accuracy, adaptability and efficiency of the proposed method for simultaneous estimation of SOC and capacity, experimental validations are performed on packs under various dynamic conditions. The results show that, for individual cells, the maximum mean absolute error (MAE) in SOC and capacity estimation is 0.92 % and 1.46 %, respectively. For the battery pack estimation, the maximum MAE is 0.96 % and 2.03 %, demonstrating the effectiveness of the proposed method.