State of Charge Estimation of Lithium-ion Batteries Electrochemical Model with Extended Kalman Filter

Abstract

Lithium-ion batteries as a dominant green energy are widely used in electrical vehicles (EV) due to their unique advantages. The battery modeling, parameter identification and state estimation are always the emphases of research which complement each other in a battery management system (BMS). Compared to the mainstream of the current equivalent circuit (EC) models, the rigorous electrochemical model with high complexity and tight coupling is not suitable for on-line simulation in EV. In this paper, the state of charge (SOC) estimation using extended Kalman filter (EKF) algorithm is proposed based on the simplified electrochemical model-single particle (SP) model. The battery parameters identified by the particle swarm optimization (PSO) algorithm show a higher accuracy, which can track the terminal voltage effectively. The SOC estimation results show that SP model with EKF algorithm is a computational method with a good performance of robust, accuracy and stability which can be used in energy management systems of EV.