Research On Power Lithium Battery Modeling And SOC Estimation Strategy

Electric vehicles are a kind of clean and environmentally friendly transportation that can effectively alleviate environmental pollution and energy crisis.For the current imperfect performance of the power lithium battery system,based on the theory of complex system modeling and analysis and advanced simulation tools,this paper carried out in-depth studies of lithium battery modeling and SOC estimation strategy,aiming at providing accurate technical support for BMS,avoiding the interference of battery “virtual power”and promoting the rapid development of electric vehicles.Firstly,in order to meet the requirement of SOC estimation for electric vehicle BMS,a dual RC equivalent circuit model is established.In view of the problem that the traditional battery model parameter RC is a fixed value and the accuracy of the model is insufficient,a variable parameter equivalent battery model based on temperature and SOC is studied.The battery was tested for pulsed discharge at different ambient temperatures,and based on the least squares identification principle of forgetting factor,the MATLAB toolbox was used to identify the parameters,then the corresponding relationship between the five parameters and the SOC in the second-order RC battery model was obtained,and the accurate modeling of the battery was achieved.Then,for the problem that measurement error and system noise covariance of the EKF algorithm is often set to fixed value during the SOC estimation,which easily causes the filter divergence and the problem that real-time battery capacity of the EV is not accurate,an AEKF algorithm was studied based on the dual RC battery model.The Kalman filter theory was analyzed,the adaptive extended Kalman filter algorithm was deduced in detail,and the simulation model of the algorithm is established.Then the initial value of the algorithm and the experimental conditions are designed.Finally,the accuracy of AEKF algorithm is verified by simulation experiments.Finally,in order to verify the SOC estimation effect of AEKF algorithm under complex current fluctuation conditions,a complete vehicle model of theelectric vehicle was built in the environment of ADVISOR software.The model parameters of each module were designed,and a half-physical simulation verification of the algorithm was performed under the UDDS operating conditions.In addition,based on the laboratory BMS experimental platform,the lithium battery was tested for charging and discharging.Through comparison of the experimental data collection and simulation in the test process,it was verified that the AEKF algorithm has higher accuracy under relatively complex working conditions,and has a good applicability in the electric vehicle BMS system.