Research On Observers For State Of Charge And State Of Health Of Lithium-ion Batteries

Recently,due to advantages of zero emission,high energy utilization and low noise features,Electric Vehicle became the important directions of automobile development.Batteries,as energy storage components of electric vehicles,have been the main bottleneck restricting the development.Compared with lead-acid batteries,nickel-hydrogen batteries and nickel-chromium batteries,lithium-ion batteries are widely used in electric vehicles due to their high energy density,fast charging speed and no memory.Battery Management System(BMS)can ensure the safe and effective operation of lithium-ion batteries.The estimation of state of charge(SoC)and state of health(SoH)of lithium-ion batteries is the basis of BMS.Therefore,how to improve the accuracy and efficiency of SoC and SoH has become a hot research topic.The establishment of high precision battery model is the basis of SoC and SOH estimation.The battery model is mainly divided into the equivalent circuit model and the electrochemical model.Due to the high accuracy of the latter model,this paper using it to estimate SoC and SoH.The electrochemical model of average electrode is a widely used electrochemical model.Based on the analysis of the average electrode electrochemical model,the average electrode electrochemical model is modified,and the electrochemical model in terms of charging and discharging conditions is established.Compared with the equivalent circuit model,the proposed model is verified by the experimental data obtained from the battery test scheme,which designed by the United States Advanced Battery Consortium(USABC).Based on the electrochemical model,a high-order sliding mode observer is proposed to estimate SoC,which solves the problem that Luenberger sliding mode observer errors can only converge to the neighborhood in theory,resulting in large observation errors and discontinuous control signal,leading to large chattering.The accuracy and robustness of SoC estimation by high order sliding mode observer are verified from USABC test data.This paper chooses the internal electrochemical parameters as indicator of the SoH,i.e.the battery capacity and the maximum concentration of lithium-ion,and establishes a sliding mode observer for estimating the SoH by using the characteristics which represent the health status of the lithium ion battery change slowly.In this paper,a non-singular terminal sliding mode observer is proposed to estimate SoH,which solves the problem of slow convergence and large error ofSuper-twisting sliding mode observer in estimating SoH.The accuracy and robustness of the proposed non-singular terminal sliding mode observer for SoH estimation are verified by the data obtained from USABC test experiments.The estimated error is less than 5%,which meets the actual requirements.The methods proposed in this paper,can be used not only to estimate the internal state of batteries,but also be designed some rules to guide the charging of batteries rapidly and estimate the remaining mileage of vehicles.It is significant to the development of battery management system.