A Study On Lithium Battery State Of Charge Estimation

Entering the 21st century,issues such as the energy crisis and environmental pollution have become more prominent globally.Under such circumstances,electric vehicles have drawn great attention from major automobile manufacturers in various countries in the world.Battery Management System(BMS)as an important part of electric vehicles,its function was to effectively manage,control,detect and protect the battery system to ensure the efficient use of battery and traffic safety.In the BMS,the key and difficult point of the research was how to improve the estimated state of charge(SOC)of the battery and further improve the driving range of the electric vehicle.The research object of this thesis was selected as the state of charge of the monolithic lithium iron phosphate battery.The battery model and the model.based SOC estimation method were studied in depth.First of all,based on the working principle,working characteristic parameters and parameter characteristics of lithium iron phosphate battery,a second.order Thevenin equivalent circuit model considering the variation of battery capacity was proposed.The amount that needs to be identified in the model includes battery capacity,open circuit voltage,battery internal resistance,and polarization capacitance.In order to realize the parameter identification of the model,an experimental platform was set up,and the actual usable capacity of the battery,the open circuit voltage and the internal resistance of DC were identified first.On the basis of this,HPPC was used in combination with the system identification toolbox in Matlab to identify the ohmic resistance,the polarization resistance and the polarization capacitance in the equivalent circuit model.Battery model parameters were fulfilled at each SOC state point.In order to verify the validity of the parameters,a simulation model was set up in Simulink.The equivalent circuit model was used in combination with the safety integration method to simulate.The simulation condition was set to constant temperature and constant current discharge with variable temperature pulse discharge.The results were compared with the model simulations.The comparison results showed that after the parameters were refined,the estimation results of the integration method could be effectively improved,but there was still a problem that the accumulated errors became larger in the later stage of simulation.Then,the disadvantages of the traditional security integration method were analyzed,and Kalman filtering was chosen to correct it.According to the setting of equivalent circuit model and the result of parameter identification,the state variables,state space equation and state observing equation needed in the estimation process were determined,and the simulation model was established in Simulink.The experiments of constant temperature and constant current discharge,variable temperature and variable current discharge,fast discharge current at variable temperature were carried out.The experimental results were compared with the simulation results.The anti.interference ability of the extended Kalman filter and its convergence to true The speed of value,and robustness in a rapidly changing environment.The results showed that the extended Kalman filter combined with the model could control the SOC estimation accuracy of single cells better than 2%,and the convergence speed of the algorithm was fast,the anti.interference ability was strong,the robustness was good,and the cumulative error was better small.