The Study On SOC Estimation Of Ternary Lithium Battery Based On Unscented Kalman Filter Algorithm(UKF)

Battery Management System(BMS)is the guarantee of the power batteries efficient and safe work in electric vehicles,and is the prerequisite for reliable operation of electric vehicles.The basic functions of BMS include battery state monitoring and State of Charge(SOC)Estimation,battery safety protection,battery energy management and so on.Among the many functions of BMS,because the estimation of the SOC depends on the related monitoring results of the battery,it makes it difficult to estimate the exact state of charge.SOC not only serves as a judgment basis for strategic design of other functions in BMS,such as energy management,but also plays a crucial role in vehicle control of electric vehicles.Therefore,accurate estimation of the state of charge is currently a hotspot of research in this field.This paper bases on the estimating of the SOC,and has carried out the following work:In order to describe the electrical characteristics of the ternary lithium battery as fully as possible and consider the calculation,the second-order RC equivalent circuit model was selected to model the battery,perform intermittent charge and discharge test and Hybrid Pulse Power Characteristic(HPPC)on the battery.Based on the experimental results,analyze the relevant characteristics of the lithium-ion battery and perform parameter identification,establish the battery model in Matlab / simulink and carry out the simulation test.According to the established second-order RC equivalent circuit model,The Unscented Kalman Filter(UKF)with system noise adaptive matching is used to estimate SOC,and write algorithm program in Matlab to verify accuracy under the working conditions.The UKF algorithm under constant current charging and discharging condition,Dynamic Stress Test condition and Urban Dynamometer Driving Schedule condition is selected for simulation test.Simulation results show that under constant current charging and discharging conditions,the accuracy of UKF algorithm with noise adaptive matching module is higher than that of UKF algorithm,and the overall error of improved UKF algorithm can be controlled within4%;The overall error of the improved UKF algorithmcan be controlled within 3.6% under DST operating conditions,the error can be controlled within 4.76% under UDDS operating conditions,so the UKF algorithm with noise matching module has high estimation accuracy.The UKF convergence verification test shows that the improved UKF algorithm has a good error-correcting ability and can quickly converge the SOC to the actual value when the initial value is not accurate.