Detection and quantitative diagnosis of micro-short-circuit faults in lithium-ion battery packs considering cell inconsistency

Micro short circuit (MSC) fault diagnosis is thought functional in preventing thermal runaway of lithium-ion battery packs. Inconsistencies in the initial state-of-charge and aging state inevitably exist among cells of a battery pack. The existing method for MSC diagnosis disregards the symptoms originating from cell-to-cell inconsistency, which may lead to misdiagnosing inconsistent cells as MSC cells and vice versa. This work presents a method for detecting and quantitatively diagnosing MSC faults in lithium-ion battery packs, while taking cell inconsistency into consideration. Initially, the median incremental capacity (IC), derived based on ranking the terminal voltages of cells, is used as a benchmark representing the state of normal cells. Subsequently, the correlation coefficients between the ICs of individual cells and their median IC are calculated in both the time and frequency domains, as to distinguish the normal, inconsistent, and MSC cells. After detecting the MSC cell, an algorithm, which is based on a recursive least squares algorithm with forgetting factor and an adaptive H_∞ Kalman filtering, is designed to calculate the short-circuit resistance online. The experimental results demonstrate that the short-circuit resistance estimated by the proposed algorithm exhibits rapid convergence to the actual values, thereby confirming the utility of the proposed algorithm in real-life contexts.