Impedance Based Dynamic Modelling,Full-frequency Characterization And Quick Diagnosis Of Lithium-Ion Batteries

Due to high energy density,long useful life and low self-discharge rate,lithium-ion batteries(LIBs)are widely used in portable electronic devices,electric vehicles and static energy storage systems.Safe and reliable operation of LIBs necessitates advanced modelling,characterization and diagnosis methods.To achieve these goals,the in-depth understanding of polarization processes of LIBs is necessary.However,current understanding toward the polarization processes is yet vague and qualitative,leaving separation of polarization losses and quantification of multiple contributions to polarization growth a challenging gap.Based on electrochemical impedance spectroscopy(EIS)technique,this dissertation focuses on the dynamic modeling,polarization processes analysis,impedance growth analysis and on-line diagnosis of LIBs with an in-depth understanding of polarization processes.The main contributions of this work is as follows:Firstly,the impedance-based dynamic modelling method is proposed.For dynamic modelling,the impedance data are used.To ensure the simplicity and practicability,the first-order RC equivalent circuit model(ECM)is used.And the impedance response of the model in the full frequency domain is derived.Then,the parameterization of the model is performed using only the low-frequency diffusion impedance data.And the obtained model can accurately describe the dynamic voltage response sampled at one second interval.Compared with the traditional time-domain parameterization method,the main advantage of this method is that the obtained ECM has physical interpretation and the simulation accuracy is higher.Secondly,the methods for separation and quantification of full-frequency polarization processes are proposed.The distribution of relaxation times(DRT)method is employed to deconvolute the impedance data in a wide frequency range of LIBs based on the appropriate modeling of the divergency of low-frequency impedance.Different polarization processes are separated.Furthermore,the EIS data are decomposed,and the impedance response of different polarization processes in the frequency domain are described.On the other hand,under the constant current discharge pulse,quantitative decomposition of polarization voltage losses is achieved.The voltage responses of different polarization processes in the time domain are clearly presented.Moreover,the polarization resistance of each process and direct current resistance under different pulse duration are analyzed.The contributions of different polarization resistances to the overall polarization are quantified.And the polarization components of the direct current resistance under different pulse duration are analyzed.Thirdly,the causes for impedance increase of LIBs aged by cycling at 45~oC are quantitatively analyzed.Based on the DRT method,the impedance data of the aged battery are analyzed,revealing that the low-frequency diffusion impedance is the largest sole source to the impedance growth.Furthermore,an advanced physics-based impedance model is applied to distinguish the solid-phase and electrolyte-phase diffusion process.And the electrolyte-phase diffusion dominates over the solid-phase diffusion in the low-frequency diffusion impedance.The increase of diffusion impedance is caused by the decrease of electrolyte diffusivity.Finally,a multi-point impedance diagnosis technique that is easy to complement,simple to calculate is proposed for monitoring and screening of lithium ion batteries.With impedance measurement on three characteristic frequency points which are calibrated using the DRT method,this technique can monitor ohmic,contact and solid electrolyte interphase resistances in real time.