Lithium-Ion Battery Performance Evaluation Based On Time-Domain Impedance Spectroscopy Measurement

With the rapid development of new energy vehicle industry,the recovery and utilization of decommissioned lithium-ion batteries has attracted wide attention.The cascade utilization is an effective way to solve the problem of recovery and utilization of decommissioned lithium-ion batteries.It is very important to evaluate the performance of lithium-ion batteries before they enter the cascade utilization.A time-domain impedance spectroscopy measurement platform for lithium-ion batteries is designed and built.A normalization method for impedance spectroscopy of lithium-ion batteries under different state of charge(SOC)is proposed.On this basis,the accurate estimation of state of power(SOP)and state of health(SOH)of lithium-ion batteries is realized.The specific research work is as follows:Firstly,aiming at the measurement of impedance spectra of lithium-ion batteries,a time-domain impedance spectroscopy measurement platform is designed and constructed.Voltage excitation signals with frequency series of quasi-logarithmic intervals are generated in the frequency range of 0.1Hz~1kHz.VI conversion circuit is designed to convert it into current excitation signals acting on the batteries to be measured,and current excitation and voltage response signals of actual batteries are collected,Fast Fourier transform is used to measure the impedance spectroscopy of batteries,the repeatability and accuracy of the measurement platform are verified.Secondly,aiming at the different influence of SOC on the size and shape of impedance spectra of lithium-ion batteries,an equivalent circuit model of impedance spectra of lithium-ion batteries is established.The parameters of the model are identified by complex variable non-linear least squares regression method,and the accuracy of the model and the effect of the identification method are verified.The relationship between the impedance spectroscopy and SOC is studied,and the sensitivity of the Four highly sensitive parameters were selected,and the key parameters of impedance spectra of lithium ion batteries were normalized by data fitting.Finally,aiming at the problem of SOP and SOH estimation of lithium-ion batteries.With fractional calculus as a tool,the end-voltage response of impedance spectrum equivalent circuit model in time domain is simulated and calculated to estimate SOP.On the other hand,the relationship between parameters of impedance spectrum model and SOH is studied,and the BP neural network is established.The experimental data of cyclic aging are used to train the neural network,and the trained neural network is used to estimate SOH,and the estimation accuracy of SOP and SOH is verified respectively.In this paper,the impedance spectroscopy of lithium-ion batteries are taken as the research focus.The problems of time-domain impedance spectroscopy measurement oflithium-ion batteries,the normalization of impedance spectra of different SOC lithium-ion batteries,the estimation of SOP and SOH based on impedance spectroscopy are solved in turn.The performance evaluation of lithium-ion batteries based on time-domain impedance spectroscopy measurement is realized,which has certain application value for the cascade utilization.