A Method of Lithium-Ion Battery Failure Diagnosis Based on Parameter Boundaries of Heterogeneous Multi-Physics Aging Model

Yaxuan Wang,Junfu Li,Shilong Guo,Ming Zhao,Weiwei Cui,Lianfeng Li,Lei Zhao,Zhenbo Wang
DOI: https://doi.org/10.1016/j.jpowsour.2023.233235
IF: 9.2
2023-01-01
Journal of Power Sources
Abstract:Lithium-ion batteries experience capacity fading during long-term operation, and they need to be replaced when reach a certain aging degree. However, it is difficult to measure a complete charge/discharge curve to obtain the capacity whether the battery is used for vehicle or energy storage. The aging state of the battery can only be judged according to the partial data of voltage and current based on the data-driven method, which is not reliable enough. In this work, a new method of battery failure diagnosis in terms of capacity fading is proposed based on the heterogeneous multi-physics aging model of lithium-ion batteries. The key parameters are obtained by parameter identification method, and the parameter boundaries when the battery is on the verge of failure are obtained by model driven method. In practical application of batteries, the failure of batteries can be diagnosed online by monitoring the key parameters, so that an early warning signal can be provided when the battery reaches the end of its life, which is of great significance to ensure that the battery can meet the performance and safety requirements.
What problem does this paper attempt to address?