Thermal runaway feature extraction and risk prediction in electric vehicle battery packs under high-temperature

TY - JOUR

T1 - Thermal runaway feature extraction and risk prediction in electric vehicle battery packs under high-temperature

AU - Yang, Haixu

AU - Liu, Peng

AU - Chen, Saihan

AU - Wang, Qing

AU - He, Yipeng

AU - Zhang, Zhaosheng

N1 - Publisher Copyright: © 2026 Institution of Chemical Engineers. Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

PY - 2026/5/1

Y1 - 2026/5/1

N2 - Thermal runaway is a critical issue constraining the development of batteries and electric vehicles. Hot weather, high-power operation, and fast charging can easily cause battery temperatures to rise, intensifying internal reactions and reducing stability. Investigating battery thermal runaway solely at room temperature fails to meet safety requirements. This study focuses on the thermal runaway behavior of battery packs under high-temperature conditions. By triggering a short circuit within the battery pack via needle penetration, temperature and voltage characteristic data during thermal runaway are obtained. Based on this data, a three-level thermal runaway warning strategy is established. Based on the evolution of characteristic data, the propagation path within the pack is identified to guide structural design. Using dynamic time warping, similarity calculations are performed between electric vehicle battery operational data and thermal runaway characteristic voltage data. Highly similar real-world vehicle data is deemed to pose safety risks. Comparisons between computational results for accident vehicles and normal vehicles demonstrate the feasibility of this approach. Under the combined effects of high temperature, pressure, and ejected particles, arc breakdown is a common phenomenon during thermal runaway in battery packs. Electrical simulation studies are conducted on three-dimensional models of the pack, quantifying the electric field strength under various conditions and assessing the hazards associated with arc breakdown. This study proposes a thermal runaway containment method from a structural and data-driven perspective, which can provide guidance for battery pack safety design. The achievement can reduce the public's concern about the safety of electric vehicles, increase the utilization rate of electric vehicles, and contribute to zero-carbon transportation.

AB - Thermal runaway is a critical issue constraining the development of batteries and electric vehicles. Hot weather, high-power operation, and fast charging can easily cause battery temperatures to rise, intensifying internal reactions and reducing stability. Investigating battery thermal runaway solely at room temperature fails to meet safety requirements. This study focuses on the thermal runaway behavior of battery packs under high-temperature conditions. By triggering a short circuit within the battery pack via needle penetration, temperature and voltage characteristic data during thermal runaway are obtained. Based on this data, a three-level thermal runaway warning strategy is established. Based on the evolution of characteristic data, the propagation path within the pack is identified to guide structural design. Using dynamic time warping, similarity calculations are performed between electric vehicle battery operational data and thermal runaway characteristic voltage data. Highly similar real-world vehicle data is deemed to pose safety risks. Comparisons between computational results for accident vehicles and normal vehicles demonstrate the feasibility of this approach. Under the combined effects of high temperature, pressure, and ejected particles, arc breakdown is a common phenomenon during thermal runaway in battery packs. Electrical simulation studies are conducted on three-dimensional models of the pack, quantifying the electric field strength under various conditions and assessing the hazards associated with arc breakdown. This study proposes a thermal runaway containment method from a structural and data-driven perspective, which can provide guidance for battery pack safety design. The achievement can reduce the public's concern about the safety of electric vehicles, increase the utilization rate of electric vehicles, and contribute to zero-carbon transportation.

KW - Arc breakdown

KW - Electric vehicles

KW - Safety warning

KW - Thermal runaway

KW - high-temperature condition

UR - https://www.scopus.com/pages/publications/105034624120

U2 - 10.1016/j.psep.2026.108782

DO - 10.1016/j.psep.2026.108782

M3 - Article

AN - SCOPUS:105034624120

SN - 0957-5820

VL - 211

JO - Process Safety and Environmental Protection

JF - Process Safety and Environmental Protection

M1 - 108782

ER -