Estimation of state of charge of lithium battery based on BCRLS-UKF algorithm

Abstract

Accurate and rapid estimation of the state of charge (SOC) of lithium batteries is one of the core functions of the battery management system (BMS). Model parameter identification is the prerequisite for accurate SOC estimation. In order to improve the accuracy of SOC estimation, a second-order RC equivalent circuit model is selected. In order to effectively avoid the influence of noise on parameter identification, the deviation compensation recursive least square method (BCRLS) is used for online parameter identification. Firstly, the principle of the Unscented Kalman (UKF) algorithm is analyzed. Secondly, the two algorithms are combined to estimate the SOC of the lithium ion battery. Finally, the data of the lithium battery SOC is estimated by comparing the two algorithms of BCRLS-UKF and UKF. The research results show that: Compared with the UKF algorithm, the BCRLS-UKF algorithm can quickly converge the initial value error, and the maximum error in the steady state does not exceed 2.5%, which verifies the effectiveness of the algorithm and robustness to external interference. It can be used In order to achieve an accurate estimation of the state of the lithium-ion battery.