Research And Realization On SOC Estimation For Lithium Ion Battery

With the rapid development of the world economy,the energy resources depletion and environment deterioration are two major challenges facing the world today.Electric vehicles(EVs)have found wide applications since they are zero-emission and flexible power configuration.For electric vehicles,the main source of power comes from the battery.How to maximize the performance of the battery on the premise of the assurance of safety is the main function of the battery management system(BMS),while the state of charge(SOC)of the battery is one of the most important functions to BMS.SOC estimation not only shows the remained capacity of the battery,but also is the prerequisite of battery equalization control and energy management.Accurate SOC prediction is of great significance to BMS.Prevention at the lithium-ion level,aim at the research of forecasting algorithm of SOC,the main work of this paper is as following:1.Research on the SOC estimation based on extended Kalman filter(EKF)algorithm and particle filter(PF)algorithm.A variety of factors which can affect the estimation of SOC were analyzed.Expound the theories and the methods to obtain the related parameters of two equivalent circuit models,Thevenin first order RC model and the combined battery model.The EKF and PF methods are combined with this two battery models respectively to establish the state space equation under the fundamental of AH counting method,and realize the more accurate estimation of SOC.2.An experimental platform for studying the estimation of SOC has been set up.The test battery is M7568215,together with AD7280,Hall sensor and MS9S12XS128 and other peripheral devices to build the test bench to sample the battery information.Such as temperature,voltage and current.The real-time monitoring and storage ofsampled data through the CAN bus to communicate with the MCU with the host computer designed by LabVIEW are realized.3.A method of SOC estimation based on EKF is designed,and the NEDC process is simulated on the test platform.And the experiment data is dealing off-line with PF.The results show that EKF and PF can effectively improve the estimation accuracy of SOC compared with AH counting method.Combined with the experimental results,the causes of errors were analyzed in detail.