Experimental Study On Aging Of Lithium-ion Battery Based On Acoustic Emission

Acoustic emission technology has been used to study the electrochemical process of battery for decades,but it is mainly used to detect the characteristics of battery electrode materials,and acoustic emission technology is less used to predict battery capacity.In this thesis,the acoustic emission technology is used to study the battery aging test,analyze and process the acoustic emission signals under different charge discharge rates,cyclic aging process and different temperatures,and finally realize the battery capacity prediction.Firstly,the working principle of lithium battery and the generation principle of battery electrode acoustic emission signal are described,which shows that the intercalation / de lithium in electrode active materials is the main acoustic emission source.Then,the lithium battery detection platform based on acoustic emission and the design of acoustic emission experimental system are introduced.The special feature of the acoustic emission signal acquisition setting in this thesis is to set the samples after the duration,so that the acoustic emission waveform generated by lithium insertion / lithium removal of the battery can be saved in an acoustic emission signal.This thesis mainly designs three kinds of experiments: different charge discharge ratio experiment,cyclic aging experiment and temperature experiment.The battery is designed to charge and discharge at different charge and discharge rates,and the battery acoustic emission signal is collected online.Using typical parameter characteristics and various signal analysis methods to analyze the data,it is found that the amplitude symbol of the threshold crossing point of acoustic emission signal waveform is related to the movement direction of lithium ion.The peak amplitude and waveform time interval of acoustic emission signal are positively correlated with the battery charge discharge ratio.The acoustic emission signal is decomposed by wavelet packet,which proves that the acoustic emission sources of battery charge and discharge are consistent.It is found that the waveform time interval and signal time interval can be used to distinguish the constant current and constant voltage charging state of the battery.The battery capacity can be analyzed by acoustic emission signal amplitude and energy.The acoustic emission data of temperature experiment is used for battery capacity prediction.The traditional battery capacity prediction mainly uses the data of battery voltage,current,temperature and internal resistance.In this thesis,the acoustic emission characteristic parameters in temperature experiment and CNN and LSTM networks are used to predict the battery discharge capacity,which preliminarily proves the feasibility of acoustic emission technology for battery condition monitoring.By analyzing the acoustic emission signal of lithium-ion battery,this thesis establishes the relationship between the battery capacity and the characteristic parameters of acoustic emission signal,which provides a new idea for using acoustic emission technology to predict the battery capacity.