A Soft Short-Circuit Diagnosis Method for Lithium-Ion Battery Packs in Electric Vehicles

Abstract

The early detection of soft short-circuit (SC) faults in lithium-ion battery packs is critical to enhance electric vehicle safety and prevent catastrophic hazards. This article proposes a battery fault diagnosis method that achieves joint soft SC fault detection and estimation. Specifically, based on an augmented state-space battery model, an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$H_{\infty }$</tex-math></inline-formula> nonlinear observer is constructed to estimate state of charge (SOC) and soft SC current in the presence of model parameter variations. Then, the asymptotic stability of the estimation error system under the desired <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$H_\infty$</tex-math></inline-formula> performance is formally proved and a tractable observer design criterion is derived. Furthermore, a diagnosis algorithm is developed to detect soft SC faults via checking the difference between the estimated SOC from the observer and the calculated SOC from Coulomb counting. Once a soft SC fault is detected, the algorithm also allows the soft SC resistance to be calculated from the estimated soft SC current. Finally, soft SC experiments of a series-connected battery pack under different working conditions and various SC resistance values are conducted to verify the effectiveness of the proposed method.