Online State Estimation Using Particles Filters of Lithium-Ion Polymer Battery Packs for Electric Vehicle

Accurate state estimation is required for high performance of battery management system (BMS) in electric vehicle (EV). The particle filter (PF) is introduced in the process because of the nonlinear feature exists in the battery system. This paper proposes a PF-based method for estimating state of charge (SOC) based on a battery equivalent circuit model. The model is established based on battery characters and the parameters in the model are on-line identified using the recursive least square with forgetting factors. The state space model of PF is obtained from the battery model. All experimental data are collected from a real Li-polymer battery. The experimental errors of SOC estimation based on PF are less than 0.2, which confirms the good performance. Moreover, the contrastive result of PF and Extended Kalman Filter (EKF) show that PF have significantly better estimation accuracy in SOC estimation.