Early Internal Short Circuit Detection and Quantitative Assessment in Lithium-Ion Battery Packs Based on Hausdorff Distance and Charging Voltage Segment Differences

Lithium-ion batteries (LIBs) have become ubiquitous in cutting-edge energy storage solutions. Identifying early internal short circuits (ISCs) is paramount for maintaining the safe and reliable operation of LIB systems. Hence, this paper introduces an innovative approach for early ISC detection and quantitative evaluation in LIB packs, leveraging Hausdorff distance and discrepancies in charging voltage segments. The median terminal voltage profile is extracted to depict the healthy state within the LIB pack. By integrating mean normalization with moving average filtering (MAF), discrepancies in voltage among cells are enhanced. Subsequently, Hausdorff distance is employed to identify and differentiate ISC-affected cells from healthy ones. For the identified short-circuited (SC) cells, their charging voltage segments are contrasted with those of healthy ones to estimate the SC current and resistance. Experimental findings confirm that our proposal precisely identifies SC cells within LIB packs. Notably, the minimum relative error in estimating SC resistance stands at 1.40 %, underscoring the efficacy and viability.