An aging- and load-insensitive method for quantitatively detecting the battery internal-short-circuit resistance

Abstract

Early-stage detection of the battery internal short circuit (ISC) can effectively prevent batteries from fire-related accidents. However, it is technically challenging because the current leakage signals at the beginning stage of the ISC could be submerged by the joint effect of battery aging, battery balancing, measurement noise, and highly dynamic charging/discharging behaviors. Aiming at these issues, we here propose an equalization-based method to quantitatively calculate short circuit resistance for pack applications. The relationship between the balanced capacity and short circuit resistance is mathematically derived and unified for both passive and active equalization systems. Battery-in-the-loop experiments show that when the short-circuit resistance changes from 55 to 330 Ω, the error of our prediction can be well-bounded within ±2%, outperforming most existing algorithms. Further, the computational burden of the proposed method is similar to that of the Ah-counting method, requiring no additional information on the batteries’ aging status. Since our method supports arbitrary load profiles, it can be used in a wide range of applications such as electrified vehicles and renewable energy storage systems.