The State Of Charge Estimation Of The Lithium-ion Battery For Electric Vehicles

With the decrease of petroleum resources,electric vehicle—the hot vehicle model for replacing traditional fuel vehicle,is receiving the emphasis of various vehicle enterprises.By virtue of its advantages of long service life,high specific energy,and no pollution,lithium-ion battery is more suitable for being used as the power unit of electric vehicle.Due to the limited energy and capacity of lithium-ion battery,for the sake of efficient and safe work of each monomer lithium-ion battery in practical use,battery management system is inevitable.Firstly,this paper designed the battery management system with the purpose of estimating battery SOC,which can collect the voltage and total current of 4 monomer lithium-ion batteries.Meanwhile,the 4 temperature sensors in external connection can realize the passive balance of each lithium-ion battery;besides,it connects with other components through CAN communication.The advantages and disadvantages of several common battery models were analyzed so as to determine to use first-order Thevenin model to simulate the electrical characteristics of lithium-ion battery.Based on different charge-discharge current,and temperature experiment,it analyzed the charge-discharge characteristic,capacity variation,voltage characteristic and internal resistance feature of lithium-ion battery.By combining with experimental data and nonlinear least squares,it completed the determination of parameters in the model and established test model in Simulink.Through working conditions of constant-current charge as well as discharge,and the complex operating condition of vehicle running,it verified the accuracy of model parameter.By means of Thevenin model and EKF algorithm,this paper estimated battery electrical charge state,which realized the offline simulation of electrical charge state in MATLAB.Based on the working condition of constant-current discharge and the complex operating condition of vehicle running,it verified the feasibility of EKF algorithm.Later,by putting EKF algorithm into battery management system,it realized the online regulation of EKF algorithm.By comparing the SOC that practically measured in HT-V5C200D200-4 battery testing device with the SOC obtained through EKF algorithm,it concluded that,the maximum error in offline simulation was smaller than 1% while the maximum error in online experiment was smaller than 2%,which conformed to the estimation accuracy of SOC made by battery management system.