Unscented Kalman Filter-Based Battery SOC Estimation and Peak Power Prediction Method for Power Distribution of Hybrid Electric Vehicles

Abstract

State of Charge (SOC) is a key parameter for battery management and vehicle energy management. Recently used SOC estimation methods for lithium-ion battery for vehicles have problems of too simple a base model for the battery and large sampling noise in both the voltage and current signals. To improve the accuracy of SOC estimation and consider that the extended Kalman filter algorithm needs linear approximation of the system equation, the unscented Kalman filter (UKF) algorithm was used to reduce the influence of sampling noise, and an improved algorithm with better filtering effect and SOC estimation accuracy was proposed. Based on the SOC estimation and battery model, the peak power prediction method for the battery is proposed and used in the power distribution strategy for Series HEV. Considering the frequent changes in load current and sampling noise, an experiment was designed to verify the effectiveness and robustness of the algorithm. The experimental results show that the UKF algorithm and the improved UKF algorithm can achieve 6% and 1.5% estimation error. The power distribution strategy based on battery SOC estimation and peak power prediction is tested and validated.