A combined experimental and simulation approach for short circuit prediction of 18650 lithium-ion battery under mechanical abuse conditions. (December 2020)

• Record Type: Journal Article
• Authors: Sheikh, Muhammad; Elmarakbi, Ahmed; Rehman, Sheikh
• Is Part Of: Journal of energy storage. Volume 32(2020)
• Publication Date: 2020-12
• Subjects: Short circuit -- Cylindrical cells -- Concentric layers -- Numerical simulation -- Finite element analysis; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2020.101833
• Languages: English
• ISSNs: 2352-152X

Abstract

Highlights: SoC dependency on failure pattern varies with different loading conditions. Temperature change rate is a key indicator for the occurrence of short circuit. Quasi-static loading techniques are suitable to predict short circuit. Scaled layer thickness can be used to represent cell geometry for finite element analysis. Abstract: Lithium-ion batteries are considered an efficient energy source for current electric vehicles (EVs); however, the safety of these batteries is vital when it comes to large-scale deployment. Short circuit of batteries is one of the concerns as it can spread quickly within the battery module or pack if not controlled at the cell level. In this paper, single lithium-ion battery cell is investigated where mechanical abuse conditions are applied to investigate short circuits and propagation of failures due to short circuits. The numerical simulation tool LS-DYNA is used for the battery-layered model, each layer thickness is considered 0.3 mm, and concentrically layered formation is used for this purpose. An improved element size of 0.5 mm is used for steel casing and 1 mm for all other layers. A total of 27 layers are simulated in a single cell and the innermost radius is considered 1 mm. Displacement at short circuit, mean temperature at the short circuit, and mean maximum temperature change criterion are used to understand short circuit and propagation of failures. Simulation models are developed for quasi-static load analysis to understand the severity of failures, which can be used to reduce the risk of sequential failure of batteries in the battery pack.