Micro short circuit fault diagnosis in Li-ion cell

Abstract

Micro short circuits (MSCs) in lithium-ion battery cells are a critical safety concern, potentially leading to thermal runaway, internal short circuits, overheating, and battery degradation. Compared to normal cells, MSC fault cells exhibit reduced capacity with each charge-discharge cycle and an increasing state of charge (SOC) deviation over time. To differentiate normal cells from MSC fault cells, a fault diagnosis method based on remaining charge capacity (RCC) estimation is proposed. After each charge discharge cycle, the cell’s RCC is compared to a safe threshold value. The method uses the charge cell voltage curve (CCVC) of a fully charged reference cell to estimate RCC via standard CCVC hypothetical conversion. This approach’s accuracy is validated in constant power and constant current charging scenarios. MSC leakage current is calculated by incrementing RCC after each charge, and then converted to MSC resistance. A MATLAB/Simulink model of a battery pack with an MSC fault was developed to test the method across various charge cut-off voltages. The diagnostic procedure’s applicability to ageing cells, constant power, and multi-step charging is further confirmed through experiments with external resistance, enhancing MSC detection before thermal runaway becomes unmanageable.