Accelerated proximal gradient algorithm for lithium-ion battery state of charge estimation with outliers

In this paper, an accelerated proximal gradient based forgetting factor recursive least squares (APG-FFRLS) algorithm is proposed for state of charge (SOC) estimation with output outliers. First, a second-order resistance-capacitance (RC) equivalent circuit model is built to reflect the operating characteristics of the battery. Then, the APG method is applied to correct the output outliers. The FFRLS and extended Kalman filtering (EKF) are used to estimate the battery model parameters and SOC interactively. In order to verify the effectiveness of the proposed algorithm, this paper models the Samsung lithium battery and compares the effectiveness of different algorithms in estimating SOC. The experimental results show that the proposed APG-FFRLS-EKF algorithm has higher accuracy.