Research On State Of Health Estimation Method For Electric Vehicle Power Lithium-ion Battery

As the core energy of electric vehicle,the advantages and disadvantages of the power battery directly determines the overall performance of electric vehicle.In order to ensure the safety and stability of electric vehicle,it’s necessary to make the accurate and reliable estimation of power battery State of Health In this dissertation,power lithium battery is taken as the research object,and its model and SOH estimation are studied.The main research of this dissertation is as follows:(1)Based on the analysis of the internal structure and operating principles of the power lithium-ion battery,the battery characteristics test has been completed,and the internal resistance is selected as the main parameters of the battery SOH by the experiment data.The demerits of various equivalent models of power battery are analyzed and compared in detail,and combined with the characteristics of power lithium-ion battery,the second-order RC equivalent circuit of the battery is chosen clearly as the research model.According to the variation characteristics of battery model parameters,the parameter identification algorithm combining recursive algorithm and least-squares algorithm with genetic factor is selected to identify the process of parameter identification,which restrain the problem that the new data is difficult to update due to the accumulation of old data.The reasonability and effectiveness of battery model and parameter identification method are verified by simulation and experiment.(2)An improved Untraced Kalman filtering algorithm is designed to estimate the SOH in view of the nonlinearity of battery system and the affected precision by noise in the model state.By introducing the adaptive covariance matching method to dynamic adjustment of process noise and measurement of noise in the Untraced Kalman filtering algorithm,the problem of the sensitivity of the model parameters is solved,and the accurate estimation of the internal resistance in the battery model is realized.The result of the simulation and experiment prove that the improved Untraced Kalman filtering algorithm has the characteristic of high accuracy and strong convergence.(3)By building the electric vehicle model in the AVL CRUISE platform and battery simulation model in the Simulink,the simulation experiment of the electric vehicle is carried out under the cycle condition of the New European Driving Cycle European and the urban cycle of the United States.The result of the simulation shows that compared with the traditional estimation algorithm,the Adptive Untracked Kalman filtering algorithm has strong dynamic tracking ability for the battery SOH estimation,and as the same time the algorithm has high accuracy and low estimation error.So the estimation method in battery state of health is feasible.