Research On Joint Estimation Of SOC And SOH Of Lithium Battery For Electric Vehicles

With the increasing global energy crisis,new energy vehicles represented by electric vehicles have become a research hotspot in various countries.As one of the key components of electric vehicles,batteries have problems such as low energy efficiency,short driving range and attenuation of life.Has been restricting the development of electric vehicles.Therefore,how to accurately estimate the battery operating state has become the focus of research.In view of the poor accuracy of battery state estimation and the limitations of the single Kalman filter(KF)algorithm,this paper builds a battery test platform,analyzes the experimental characteristics of lithium batteries,establishes a second-order RC equivalent circuit model,and based on The test platform data verifies the accuracy of the model.The extended Kalman filter(EKF)and the unscented Kalman filter(UKF)are used to estimate the battery state of charge(SOC).Experimental verification shows that the estimation accuracy of the UKF algorithm is better than that of the EKF algorithm.On this basis,a joint algorithm with forgetting factor least squares(FFRLS)combined with EKF,a dual extended Kalman filter algorithm(DEKF)and a combined algorithm with EKF and UKF are proposed.The three joint algorithms can realize online identification of battery model parameters and battery SOC estimation.According to the functional relationship between ohmic resistance and battery state of health(SOH),the joint estimation of battery SOC and SOH can be completed.Experimental verification of two different working conditions shows that the EKF-UKF joint algorithm has better robustness and estimation accuracy when estimating battery SOC and SOH,and under different conditions of the initial value of ohmic resistance and the initial value of SOC Next,the EKF-UKF joint algorithm can eliminate the error caused by the initial value,and can quickly converge to the true value of SOC and SOH,with good convergence and accuracy.In this paper,the EKF-UKF joint algorithm is built to better solve the problems of poor battery state estimation accuracy and the limitations of a single KF algorithm,and realize the online identification of battery model parameters and the joint estimation of battery SOC and SOH,which is an electric vehicle battery management system Efficient operation provides an accurate method of battery state estimation.