Study On Lithium-ion Battery State Of Health Estimation Based On Electrochemical Mechanism

New energy vehicles are developing rapidly with the advantages of energy saving and environmental protection,and lithium-ion batteries are widely used in new energy vehicles because of their high energy density and long cycle life.However,lithium-ion batteries are poorly resistant to abuse.To avoid overcharging and overdischarging,which can affect battery life and even lead to thermal runaway and other accidents,a battery management system needs to be developed to control them.Battery state of health(SOH)estimation is one of the core functions of the battery management system.Accurate battery state of health estimation contributes to the safe utilization of the battery and further improves vehicle performance.This paper addresses the problems of unclear physical meaning of existing battery state of health estimation algorithms and the need for long time and large aging test data for algorithm development,studies the pseudo two dimension model theory and the effects of battery aging,and proposes a lithium-ion battery state of health estimation algorithm with short required test time and clear physical meaning based on the pseudo two dimension model.A modified equivalent circuit model and a state of health estimation algorithm based on the modified equivalent circuit model are proposed for the problem of large computational effort of the pseudo two dimension model.The main research of this thesis is as follows.(1)A lithium-ion battery cycle life test platform was built,and a test plan was designed and conducted with Samsung 18650 batteries as the test object,and a battery aging database was established to lay the foundation for analyzing the battery aging law and validating the state of health estimation algorithm.(2)To address the problem that the existing state of health estimation algorithm requires a large amount of battery aging test data,a state of health estimation algorithm based on a pseudo two dimension model is proposed.In order to simulate the actual battery using the pseudo two dimension model,it is necessary to make the model parameters consistent with the actual battery as much as possible.Some parameters of the battery were determined by disassembling the battery and using scanning electron microscope measurements;the accuracy of the open-circuit potential curves of the used positive and negative materials was verified by using test open-circuit voltage data;the identification of difficult-to-measure parameters was completed by using genetic algorithms;and the accuracy of the established model was verified by using test data.Based on the pseudo two dimension model theory and considering the sensitivity analysis results of the battery parameters,the parameters related to the reversible capacity of the positive electrode were selected for identification,the influence of aging on the related parameters was analyzed,and the estimation of the battery state of health was completed,and the effectiveness of the algorithm was verified by the experimental data.(3)In order to further improve the computational speed of the state of health estimation algorithm,the first-order equivalent circuit model is used to replace several complex partial differential equations in the pseudo two dimension model,and a modified equivalent circuit model is formed by combining the positive and negative material open-circuit potential curves and using a genetic algorithm for parameter identification,and the validity of the model is verified by experiments.According to the state of health estimation algorithm based on the quasi-dimensional model,the state of health estimation is completed using the parameters related to the reversible capacity of the positive electrode obtained from the parameter identification of the modified equivalent circuit model,and the validity of the algorithm is verified with experimental data.