State of charge and surface temperature estimation of lithium-ion batteries on the basis of a fractional-order equivalent circuit-thermal coupling model

The purpose of this study is to address the issues of significant temperature impact on lithium-ion batteries and the poor temperature adaptability of traditional models that fail to accurately estimate battery temperature. This study constructs a fractional-order equivalent circuit-thermal coupling model (FOECM-TCM), which is composed of a fractional-order equivalent circuit-thermal model and a dual-state lumped parameter thermal model. This model resolves the inaccuracy in the state of charge (SOC) and temperature estimation caused by nonlinearity, coupling, and time-variant parameters in lithium-ion battery systems. The parameters of the electrical and thermal models under US06 driving conditions were identified via fractional-order particle swarm optimization (PSO) and the genetic algorithm (GA). To address the mutual coupling of the SOC and temperature in lithium-ion batteries, this paper proposes a joint estimation method based on an electrothermal coupling model. This model integrates fractional-order theory with the extended Kalman filter (EKF) algorithm to develop a fractional-order extended Kalman filter (FOEKF). To validate the superiority of the FOECM-TCM model and the FOEKF, a comparative analysis is conducted with the traditional equivalent circuit-thermal coupling model. The FOECM-TCM model reduces the root mean square errors of SOC and surface temperature estimations by 55.9% and 9.4% and the mean absolute errors by 59.2% and 17.9%, respectively. These results demonstrate superior accuracy and responsiveness, indicating excellent precision and applicability.