Research On SOH Estimation Method For Lithium Ion Battery Of Electric Vehicle

In recent years,in order to reduce the consumption of petroleum resources as well as automobile exhaust pollution,China greatly supported the development of the electric vehicle industry.Under the guidance of national policies,the number of electric vehicles was increasing,battery technology and battery control technology have occupied a extremely significant position in electric vehicles.With the widespread use of electric vehicles,the deficiencies of battery management systems have gradually emerged,among which the short driving range as well as the short service life were prominent issues in electric vehicles.In order to solve the issues existing in electric vehicles,this paper predicted the state of health based on the state of charge of the battery.The research content was of great significance for improving the driving range of electric vehicles as well as extending battery life.The main research contents of this paper were as follows:(1)By comparing the feasibility of SOH definition parameter acquisition,the capacity was selected as the evaluation parameter of SOH for battery in this paper.(2)Basic performance test of battery.The working principle of lithium ion battery was analyzed.Battery performance parameters were explained.Through the construction of the battery test platform,the basic performance test of thelithium ion battery was completed,the influencing factors of the SOH were summarized.(3)Parameter identification and verification of the battery model.By analyzing the accuracy of the battery model and the computational complexity of the model,the Thevenin equivalent circuit model was chosen as the research model.In the process of model parameter identification,temperature variables were introduced,three temperature ranges were set to identify the model parameters.The rapid test approach was used to identify the relationship between the open circuit voltage and the SOC.The RC parameters in Thevenin model were identified by exponential fitting.The reliability of the parameter identification results of Thevenin model was verified by the inverse approach.(4)Establish the state equation and observation equation of Kalman filtering.The principle of Kalman filtering algorithm was expounded.The state equation and observation equation were established which were used to estimate SOH.The Kalman filtering algorithm flow was clarified.The dual-expansion Kalman filter algorithm was used to predict the change of the battery capacity.The change of SOH was determined by the change of capacity according the definition formula of SOH.Through the battery cycle test,the prediction accuracy of the dual Kalman filter algorithm was verified.(5)Due to the low prediction accuracy of the dual Kalman filter algorithm,an unscented Kalman filter algorithm was proposed to predict SOH.The unscented Kalman filter algorithm not only can solve the linearization errorissue existing in the dual Kalman filter algorithm but improve the prediction accuracy of the algorithm.The results showed,the optimization was performed on the basis of the unscented Kalman filter algorithm.System noise and observed noise can be automatically corrected with the update of state variables,which can better simulate the noise impact of the SOH in the prediction model,which can reduce the error of the SOH estimation result and improve the prediction accuracy of the SOH.