Font Size: a A A

Soh Estimation Of Lithium-ion Battery Based On Dynamic Electrochemical Impedance Spectroscopy

Posted on:2021-02-12Degree:MasterType:Thesis
Country:ChinaCandidate:Y Z RuanFull Text:PDF
GTID:2392330605475887Subject:Materials engineering
Abstract/Summary:PDF Full Text Request
The large-scale application of lithium-ion batteries has brought us convenience and accompanied by a large number of safety hazards.Therefore,the rapid and accurate estimation of state of health(SOH)of battery has become a hot and difficult point in current research.In this paper,a new dynamic electrochemical impedance spectroscopy(DEIS)test method is proposed by improving the electrochemical impedance spectroscopy(EIS).For the first time,the DEIS is used to study the SOH of the battery,combined with the BP neural network model,the online SOH can be quickly and accurately estimated.The main research contents are as follows:Firstly,the effect of temperature stress on battery degradation is studied,and battery aging experiments are designed.Three lithium-ion batteries with different SOH are fabricated to explore the internal mechanism of battery capacity degradation and the internal relationship between DEIS and SOH.Through the electrochemical performance analysis during its entire life cycle and various physical and chemical characteristics of the battery after disassembly,studying the internal mechanism of battery capacity decay at high temperature and the influence of internal changes on the impedance during battery aging,indicating that there is an inevitable internal relationship between the reduction of SOH and the increase in impedance.Secondly,the multi-frequency superposition principle is used to improve the traditional EIS measurement,and a DEIS test method is proposed.The fast algorithm of Fourier transform is used to analyze the frequency spectroscopy of the response signal to ensure the simultaneity of the multi-frequency test during working state.An experimental platform for DEIS testing is built.Through control variable experiments,the optimal conditions for DEIS testing are determined,which effectively shorts the time required for testing.Based on the optimal test conditions,the differences between DEIS and SEIS and the change rule of the DEIS of the battery under different aging paths are studied.Finally,the original parameters of DEIS are analyzed,and the correlation between each parameter and the available capacity of the battery are calculated using the grey relation analysis method to extract the characteristic parameters that can reflect the SOH of the battery.The BP neural network model is established and optimized by genetic algorithm to obtain an improved GA-BP model.The model is trained using the characteristic parameters of DEIS and battery discharge capacity.MATLAB simulation result shows that the optimized model has a faster convergence speed,and the error in estimating the SOH of the battery during the entire life cycle is within 2.9%,and the model also has a certain generalization ability.
Keywords/Search Tags:state of health, lithium-ion battery, dynamic electrochemical impedance spectroscopy, BP neural network, genetic algorithm
PDF Full Text Request
Related items