Research On Quantitative Measurement System For Lithium Plating Of Lithiumion Battery Based On Electrochemical Impedance Spectroscopy

Under the special working conditions of low temperature,high rate and overcharge,lithium plating of lithium-ion battery will lead to the deterioration of capacity and internal resistance of lithium-ion battery,and even lead to thermal runaway caused by internal short circuit,which will bring safety accidents.At present,there is a lack of an effective quantitative method for lithium plating of commercial lithium-ion batteries.Based on the electrochemical impedance spectroscopy(EIS)of lithium-ion batteries,this paper studies the rapid and nondestructive measurement of lithium plating in commercial batteries.Build the time domain measurement platform of lithium-ion battery EIS,design experiments for specific batteries and quantify lithium plating.The EIS data are modeled and the high sensitive parameters are extracted,and the mass of lithium plating is estimated by BP neural network algorithm.The effectiveness of the proposed method is verified by scanning electron microscope(SEM)images and energy dispersive spectrometer(EDS).The specific research contents are as follows:Firstly,in order to obtain the EIS data of lithium-ion battery,a time domain measurement device based on fast Fourier transform is designed and built.The part of hardware mainly includes circuit design,board configuration and device shell design,while the part of host computer mainly carries out EIS mapping.The equivalent circuit is built,and the accuracy of the device is verified by comparing the the measured results of the device with EIS theoretical calculation value of the equivalent circuit.The same equivalent circuit is measured several times continuously to verify the repeatability of the device.Several typical batteries will be tested by EIS to verify the practical application effect of the device.Secondly,taking ur14500 p lithium cobaltite battery of Sanyo company of Japan as the experimental object,the accelerated lithium plating experiment is designed.The experimental data are preprocessed to obtain the effective data of capacity attenuation and EIS,so as to preliminarily judge the effectiveness of the experimental.The capacity loss caused by lithium plating is further transformed into the quality of lithium evolution.The fractional order model of EIS is established,and the parameter identification process of the model is designed.The sensitivity calculation method of model parameters is designed,and then the highly sensitive parameters of EIS model are extracted.Finally,based on the high sensitive parameters and BP neural network algorithm,the mass of lithium plating is estimated,and the effectiveness of the algorithm is verified from the angle of absolute error and relative error.Experiments are designed to obtain batteries with different degrees of lithium plating,and the mass of lithium plating is estimated by this method.The scheme of battery disassembly and subject verification is designed.SEM / EDS will be used to obtain the scanning image and element content to verify the effectiveness of this paper qualitatively and semi quantitatively.