Research On Joint Estimation Of State Of Charge And Usable Capacity Of Lithium-ion Power Battery For Vehicles

The State of Charge(SOC)of lithium-ion power batteries is estimated to be the basic function of the Battery Management System(BMS).Charge and discharge control,safety management,balance management,and fault diagnosis are all inseparable from accurate SOC estimation.The accurate available capacity is the prerequisite for obtaining an accurate SOC value.It not only reflects the current maximum energy storage capacity of the vehicle,but is also directly related to the use environment(temperature,working conditions,etc.)of the power battery.At the same time,the capacity will also vary with the aging of the battery gradually decays,which is closely related to the service life,energy decision and safety of electric vehicles.If the estimation problem of SOC and usable capacity in BMS can be solved,the safety problem of power battery can also be effectively solved,and then the service life of power battery can be prolonged,and application and maintenance costs can be reduced.Therefore,this article focuses on the joint estimation of SOC and usable capacity of lithium-ion power batteries,and carries out the following research work:(1)In view of the unknown performance parameters of the lithium-ion power battery,first briefly introduce the working principle of the lithium-ion power battery,build a lithium-ion single battery test platform,and design the lithium-ion power battery test plan in detail.Then the basic performance test,working condition test and cycle life test of Panasonic 18650PF ternary lithium-ion battery made in Japan were conducted.Finally,system analysis is performed on specific test results,and battery performance parameters such as initial capacity,open circuit voltage(OCV),and ohmic resistance are obtained.(2)Aiming at the adaptability of the lithium-ion power battery model under variable environmental temperatures and complex operating conditions,by weighing the accuracy and complexity of the battery model,and considering the impact of environmental temperature changes on the power battery,a temperature-variable dual polarization equivalent circuit model is established.Then the parameters such as the polarization internal resistance and the polarization capacitance of the model are identified.Finally,under constant temperature and variable temperature conditions,the accuracy and robustness of the established dual-polarization equivalent circuit model were analyzed and verified based on DST,FUDS and HPPC.The results prove that the temperature-varying dual-polarization equivalent circuit model has good accuracy,and has good temperature adaptability and working condition adaptability.(3)Aiming at the problem of lithium-ion power battery SOC estimation,the extended Kalman filter(EKF)algorithm and H_∞filtering algorithm are introduced respectively,and the basic principles and SOC estimation process of the two algorithms are introduced in detail.Then in the MATLAB/Simulink simulation software,the simulation models of the SOC estimation method based on the EKF algorithm and the SOC estimation method based on the H_∞filter algorithm are established respectively,and the two methods are applied under the conditions of constant temperature and variable temperature through three different working conditions data of DST,FUDS and HPPC.The results show that the SOC estimation method based on the H_∞filtering algorithm has higher accuracy and stronger adaptability.(4)Aiming at the SOC estimation problem under the aging behavior of lithium-ion power batteries,a lithium-ion power battery capacity attenuation model is established,and the particle filter algorithm is used to predict the battery capacity.On this basis,a SOC estimation method considering the capacity attenuation is proposed.The proposed method is analyzed and verified by using test data under six different aging and degrees.The results show that the proposed SOC estimation method considering the capacity attenuation can effectively restrain the influence of battery aging on the accuracy of SOC estimation.(5)Aiming at the problem of joint estimation of SOC and available capacity of lithium-ion power battery,a joint estimation method of SOC and available capacity based on single-time-scale double H_∞filtering is proposed,that is,two H_∞filters are used to jointly estimate the state quantity and parameter quantity of the power battery.One of the filters performs real-time estimation of power battery SOC,and the other filter performs online identification of available capacity.Compared with using only one H_∞filter,the model’s SOC estimation accuracy has been further improved.Due to the fast time-varying characteristics of the battery system state quantities and the slow time-varying characteristics of the parameter quantities,the single-time-scale double H_∞filtering algorithm is improved to a multi-time-scale double H_∞filtering algorithm,and the SOC-capacity-OCV three-dimensional response surface is introduced as the state feedback link To solve the dependence on the SOC-OCV curve,the DST working condition data after 20 cycles of charge and discharge is used for analysis and verification under normal temperature and low temperature environments.The result proves that compared with the single-time-scale double H_∞filtering algorithm,the multi-time-scale double H_∞filtering algorithm further improves the SOC estimation accuracy,which is more suitable for the actual vehicle operation.