State-of-Charge estimation for power Li-ion battery pack using Vmin-EKF

An accurate estimation of State-of-Charge (SOC) for power battery pack is very important in the applications of electrical vehicles. Single cell model is not suitable for battery pack of m u Iti cells. Considering the imbalance characteristic of serial connected battery pack, a V
m¡1¡
model is proposed. The minimal cell load voltage of the battery pack (V
m¡1¡
) is used as the model measurement variable, and SOC is used as the model state variable. Based on the V
min
state space model, the extended Kaiman filter (EFK) approach is applied to get the recursive estimation of the battery pack's SOC. Experiments were made to simulate the behaviors of battery pack in the driving conditions. The results showed that accurate and real-time estimation of SOC could be obtained through this approach.