Research On State Of Charge Estimate For Power Battery In Electric Vehicles

With the problem of energy crisis and environmental pollution becoming more serious, electric vehicles which are powered by batteries show outstanding advantages on environmental conservation. Under this situation, power battery becomes the focus of research in enterprises and colleges. In the study of power battery, state of charge(SOC) is an important factor to the performance of electric vehicle. SOC is the foundation of control strategy in electric vehicle. The lithium iron phosphate battery is taken as the research object in this paper. Then the research of SOC estimation based on the battery model is carried out.In this paper, the experimental platform is built. The method for measuring the actual capacity of battery and correction actual capacity by temperature is studied. Then through the pulse discharge experiment, the pulse response voltage curve of battery is analyzed, which is the experimental foundation for establishing battery model and identifying the model parameters.The advantages and disadvantages of some classic battery equivalent circuit models is analyzed in this paper. The PNGV equivalent circuit model is selected as the battery model of SOC estimation. The function of the model is established.Through the open circuit voltage(OCV) measurement experiments under different temperature conditions, the function between OCV and SOC is obtained. Then the influence of temperature on the OCV is analyzed. The hybrid pulse power characteristic(HPPC) experiments under different temperatures are conducted. The least square method is used to identify the model parameters based on the pulse response voltage curve. The parameters with temperature and SOC are obtained. The battery simulation model is established to prove the accuracy of the model by the custom pulse discharge experiments.The battery’s space state equations which introduce a temperature correction coefficient are established based on the PNGV model. The simulation program of SOC estimation based on the Unscented Kalman Filter(UKF) algorithm is implemented. The constant current discharge experiments and the dynamic stress test(DST) experiments are conducted. Then the accuracy of SOC estimation is proved in the consideration of initial values. The influence of initial values on the algorithm convergence is validated. The results which compared with the traditional Coulomb counting algorithm indicate that the estimation error of UKF algorithm is less than 5%. This is better than the error of Coulomb counting algorithm. The UKF algorithm has the higher accuracy and faster convergence speed in battery SOC estimation.