Research On The Thermal Runaway Early-Warning Method Of Lithium-Ion Battery Based On The Electric-Thermal-Aging Coupling Model

With the rapid development of electric vehicles and the widespread use of lithium-ion batteries,the number of thermal runaway accidents is also increasing every year,so the research of methods for thermal runaway early warning has become an important topic.The effectiveness of the thermal runaway early warning system depends on the accuracy of the parameters and methods,in comparison with the independent equivalent circuit model,the multi-scale coupling model can calculate or estimate the characteristic parameters more accurately.With the aim of providing a thermal runaway early warning system with a high level of confidence,this thesis presents an in-depth study and analysis of the coupling model and the thermal runaway early warning methods for lithium-ion batteries from 3 aspects.1.The cylindrical lithium-ion battery is taken as the research object to analyse its thermal equilibrium and ageing characteristics.The internal temperature variation gradient of the battery is approximated by radial finite difference representation,and established the onedimensional thermal model based on the heat balance principle,and establish a dual indicator battery ageing model using the actual capacity and the internal ohmic resistance as the assessment index.Combined with the dual polarisation model to analyse the battery parameters affected by thermal and ageing effects,an electric-thermal-aging coupling model was developed that can be used to estimate the operating state of lithium-ion batteries over a wide temperature range and under different life cycles.2.In order to obtain the parameters of the electric-thermal-aging coupling model and to verify its correctness,a series of experiments such as low current discharge and constant capacity pulse discharge are performed on 18650 batteries under different constant temperature conditions.The responses of the electrical model,thermal model and aging model are measured under the specific excitation,and the particle swarm algorithm and recursive least squares method are selected to calculate the best-fit parameters according to the parameter identification requirements.The accuracy of the identification parameters and the electric-thermal-aging coupling model is verified by comparison of the experimental results with the fitted curves.3.Since the existing thermal runaway early warning method is relatively single,this thesis analyses the development stages and characteristics of thermal runaway,and transforms the thermal runaway early warning into an internal short-circuit detection problem.Based on the electric-thermal-aging coupling model and the characteristics of the internal short circuit,four methods of thermal runaway warning are designed and optimised.To avoid the effects of other battery failures in addition to the thermal runaway early warning,the tree analysis method and fault signal superposition algorithm are adopted to make the four warning signals work together.Finally,the thermal runaway scenario is simulated using MATLAB/Simulink.The results show that the thermal runaway warning method investigated in this thesis can accurately detect thermal runaway and has high reliability.