Lithium-ion Battery Lifetime Prediction: Integrating Statistical Models with Physics-of-Failure Mechanisms



Journal Title

Journal ISSN

Volume Title



One of the critical concerns for lithium-ion batteries that have been widely used in the electronics and automotive industries is the evaluation and prediction of battery lifetime. Several approaches for lifetime prediction have been developed including electrochemical models, equivalent circuit based models, empirical models, and performance-based models. However, statistical models based on physics-of-failure of lithium-ion batteries have not been well established. This research begins with the analysis of the battery aging mechanisms using the Failure Mode and Effect Analysis (FMEA) method, and identifies major physical degradation models for lifetime prediction. Then a statistical method is developed to model the battery performance degradation that is induced by aging mechanisms. The nonhomogeneous Wiener process model integrated with physical degradation mechanisms, namely, capacity decay and power fading, is developed for battery degradation data analysis. The proposed model is first demonstrated through a simulation study, followed by the evaluation using real data of lithium-ion batteries. This effort shows the effectiveness of our proposed methodology of integrating statistical models with physics-of-failure mechanisms.



Battery lifetime, Lithium-ion batteries (LIB), Physics-of-failure, Statistical models