A novel collaborative multiscale weighting factor-adaptive Kalman filtering method for the time-varying whole-life-cycle state of charge estimation of lithium-ion batteries

Accurate state of charge (SOC) estimation is essential for the whole-life-cycle safety guarantee and protection of lithium-ion batteries, which is quite difficult to realize. In this study, a novel weighting factor-adaptive Kalman filtering (WF-AKF) method is proposed for the accurate estimation of SOC with a collaborative model for parameter identification. An improved bipartite electrical equivalent circuit (BEEC) model is constructed to describe the dynamic characteristics combined with the mathematical correction of the time-varying factors. The model parameters are identified online, corresponding to various SOC levels and temperature conditions. Considering the internal resistances, ambient temperature, and complex current rate variations, an adaptive multi-time scale iterative calculation model is constructed and combined with the real-time estimation and correction strategies. The maximum closed-circuit voltage (CCV) traction error is 0.36% and 0.24% for the main pulse-current charging and discharging processes, respectively. The proposed WF-AKF algorithm stabilizes the large initial SOC estimation error by tracking the actual value with a maximum error of 0.46% under the complex working condition. The SOC estimation is accurate and robust to the time-varying characteristics and working conditions even when the initial error is large, providing a safety protection theory for lithium-ion batteries.