An optimized quantum particle swarm optimization–extended Kalman filter algorithm for the online state of charge estimation of high-capacity lithium-ion batteries under varying temperature conditions

The core focus of the battery management system (BMS) is accurate state of charge (SOC) estimation of the lithium-ion batteries. To solve the problem of improper selection of the noise covariance matrix in the extended Kalman filter (EKF) algorithm, which in turn affects the actual operating effect and range of electric vehicles, this paper proposes the adaptive sine cosine–Levy flight–quantum particle swarm optimization (ASL-QPSO) algorithm to find the optimal noise covariance matrix. Firstly, this paper proposes the variable forgetting factor recursive least square (VFFRLS) algorithm to identify the parameters of the equivalent circuit model of the power lithium-ion batteries. Then, the obtained parameters are transmitted online by the EKF algorithm, based on which the local attraction factor is updated using the ASL-QPSO, which is used to select the appropriate noise covariance matrix. Finally, the optimized noise covariance matrix is obtained and used to achieve the accurate SOC estimation of the power lithium-ion batteries. Experimental results under different operating conditions and temperatures show that the maximum absolute error (MAX), mean absolute error (MAE), and root mean square error (RMSE) of the algorithm are less than 1.82%, 0.59%, and 0.72%, respectively. This demonstrates that the algorithm has superior convergence tuning and high robustness, presenting a novel optimization strategy for the SOC estimation of lithium-ion batteries.