Research on SOC Estimation of Lithium Batteries Based on Novel Fusion Algorithm

Abstract

In order to better estimate the State of Charge (SOC) of lithium batteries, this paper proposed a novel approach that combined the open circuit voltage (OCV) scheme, the ampere-hour (AH) integration strategy and the extended Kalman filter (EKF) method. Based on experimental data of battery pulse charging and discharging under hybrid pulse power, the equivalent circuit model of the second-order Resistor-Capacitance (SoRC) network was created. Besides, the curve of the corresponding relationship between SOC and open circuit voltage was fitted so as to identify the equivalent circuit model parameters of lithium batteries. Moreover, the novel SOC fusion algorithm was evaluated and simulated in MATLAB software. The obtained results demonstrated that under the dynamic test condition, the proposed fusion strategy accelerated the convergence of the EKF method for the predicted value and avoided the accumulative error of the AH integration strategy in the SOC value range of 90%-100% by utilizing the OCV to obtain the initial value of SOC. The proposed method estimates the SOC of the battery in real time and controls the SOC estimation error within 2%.