Estimation of battery open-circuit voltage and state of charge based on dynamic matrix control - extended Kalman filter algorithm

Abstract

State of Charge (SOC) estimation is one of the most important functions of the battery management system for new energy vehicles. Extended Kalman Filter (EKF) algorithm has been widely used in SOC estimation of lithium-ion batteries (LiBs). However, the model parameters in the SOC estimation algorithm change with the aging of the battery, which makes EKF unable to obtain accurate estimation results. Because of these defects, a battery model is built based on the second-order RC equivalent circuit model. The parameters of the LiB model are identified by an off-line test and the battery simulation model is calibrated. Then, based on the simulation model, the influence of Open Circuit Voltage (OCV) deviation on SOC estimation accuracy of the EKF algorithm is quantitatively analyzed. The results show that the deviation of OCV will affect the accuracy of SOC estimation by the EKF algorithm. When the OCV deviation reaches 15 mV, the SOC estimation error will reach 5%. Subsequently, a method of updating model parameters based on the Dynamic Matrix Control (DMC) algorithm is proposed. And the DMC-EKF algorithm is used to estimate OCV and SOC. The results show that after the DMC algorithm is used to linearize the RC network, the identified OCV parameter deviation is <10 mV, and the SOC deviation estimated by the DMC- EKF algorithm is <5%, which can meet the application requirements. The EKF algorithm can estimate the SOC more accurately after updating the OCV by the DMC algorithm. Compared with EKF algorithm and UKF algorithm without online OCV updating, DMC-EKF algorithm reduces the maximum deviation of SOC estimation by at least 2%. The proposed DMC-EKF algorithm has good accuracy and robustness.