Study On Lithium-ion Battery Deformation And Failure Based On Detailed Modeling

Electric vehicles have brought new challenges toward crash safety design.Large mechanical deformation and failure of the battery structure is likely to induce the internal short circuit and subsequent thermal runaway.In this thesis,the failure mechanism of lithium-ion batteries under mechanical loading is thoroughly studied through experimental design,analytical method,and finite element analysis.The results benefit a better understanding of the mechanical failure mechanism of the battery structure and facilitate the safety design of electric vehicles.Mechanical indentation tests on commercial lithium-ion batteries are performed under different loading conditions.It is found that the internal short circuit(ISC)occurs during the mechanical damage and fracture process.In most loading cases,ISC initiates simultaneously with the global fracture(local force peak)of the battery.Two fracture modes,namely the in-plane local fracture and the inter-layers' shear fracture,are observed.It is found that the open circuit voltage(OCV)response after the ISC is highly related to the fracture modes.Inter-layers' shear fracture is more likely to induce the ISC and a larger number of layers with inter-layers' shear fracture results in severer ISC and thus a faster voltage drop.The fracture modes also affect the temperature field of the battery by determining the ISC position and area.To further study the fracture mechanism of the pouch cell,the component materials were tested and calibrated,especially the coating materials,for which the Drucker-Prager/Cap(DPC)model was adopted.The damage accumulation phenomenon was observed during the compression of stacked anodes,which indicates the damage of the graphite coating.The DPC model is then modified to include this factor.A detailed multi-layer model for the pouch cell was developed and successfully predicts the force-displacement response and the two fracture modes.The fracture sequence of battery components was analyzed.The current collectors fracture first and lead to the fracture of coatings,and then the separator fails.Lithium-ion batteries usually operate at a high state-of-charge(SOC),thus we explored the influence of SOC.Mechanical indentation tests on the cylindrical and pouch cell were conducted under different SOCs.Results show that the cylindrical cell has significant SOC-dependent mechanical response and ISC behavior,however,the behaviors of pouch cell are almost SOC-independent.To explore the mechanism of this difference,two sets of tests were designed.Based on the control test results,we conclude that the internal stress to resist the charging-induced volume expansion of jellyroll is the major reason for the SOC-dependence of the mechanical behavior of lithium-ion battery cells.We also conclude that the difference in SOC-dependence between the cylindrical cells and the pouch cells is closely associated with the casing stiffness(stiff metallic casing vs.soft pouch)and the construction method of jellyroll(winding versus stacking),which together create different conditions of resisting the volume expansion and retaining the internal stress.A data-driven machine-learning model was developed to predict the failure state of battery with high efficiency and certain accuracy under impact loading.The detailed model was used to generate a large amount of data through simulation.Machine-learning models were then trained and tested with the data and can reasonably predict the failure state,failure displacement,and failure force of batteries subjected to impact loading.