State Of Charge Accurate Estimation Based On Kalman Filtering For Power Battery

Because of oil resources shortage and air quality degradation, electric vehicle with advantages of energy saving and environment protection has emerged as the main trend in automobile industry. As the major energy carrier and power source, battery's manufacturing process and group application technology have been the key factors in promoting the commercial progress of electric vehicle. Therefore, in order to keep power battery work securely and effectively, electric vehicle must be equipped with a specific management system to control and supervise the function of battery. The state of charge SOC estimation has always been the core component in battery management system, which is one of the main parameters to reflect the battery working states and can provide judgment basis to vehicle control strategy. This paper takes the LiFeO4 polymer power battery as the research object, and uses Kalman filter correction algorithm for battery pack online SOC estimation.Firstly, the paper describes the development of electric vehicle and the performance requirements of vehicular power battery, and it takes the electrochemical characteristics of LiFeO4 battery as a starting point to analyze the various SOC effect factors and study the difficulties of online accurate estimation. After comparing some commonly used methods and considering the electric vehicle environment, this paper proposes a new method named Kalman filter correction algorithm on the basis of SOC definition. The algorithm gives such full play to the advantages of open circuit voltage method, ampere hour counting method and extended Kalman fitering algorithm that makes the estimation accuracy and real-time ability improved significantly. According to discharging experimental data, the system has established discharging or charging rate model, temperature model, open circuit voltage model and extended Kalman filtering combined model.Secondly, the paper builds up the power battery pack's SOC estimation system in view of software and hardware design. This system is made up of data sampling, algorithm execution, communication management, protection control, information storage and display. It possesses so many functions that can take an online accurate estimation on vehicular battery pack in a real sense, including battery data online detection, Kalman filtering correction algorithm implementation, microcontroller communication, liquid crystal display, battery diagnosis and protection.Finally, the paper designs the battery pack's charging and discharge experiments. With the help of automobile driving cycle, such as HWFET, UDDS, FUDS, the performance of SOC estimation algorithm has been verified and improved. Consequently, with a good estimation effect, Kalman filtering correction algorithm can be completely in conformity with the SOC accuracy requirements in electric vehicle environment.