Get started
Work

An Exploratory Review on Lithium-Ion Battery Fire Risk Mitigation Using Deep Learning Approaches

Year: 2024
Type: conference-paper
Abstract: : Lithium-ion batteries (LIBs) have been extensively adopted in various applications ranging from portable electronics to electric vehicles owing to their high energy density and efficiency. Meanwhile, LIB fire risks have also increased, with many incidents causing life and property losses. With the advancement of machine learning techniques, deep learning approaches have been used in research on battery fire safety, particularly in battery safety risk prediction, risk classification, thermal runaway detection, and system optimization. This review summarizes deep learning approaches applied to mitigating LIB fire risk. As the LIB industry expands and safety regulations are relatively new, significant opportunities for improvement and unexplored technology areas remain. These include thermally stable components, smart material design, battery thermal management systems for safety monitoring, safety design models for battery systems, battery safety evaluation systems, and more. Deep learning can effectively and flexibly align with these developments. We compare and discuss state-of-the-art studies that implement deep learning in battery thermal performance and risk mitigation, especially using images for early battery fire detection. We also propose further prospects for developing new energy materials and advancing battery technology. By applying deep learning approaches, researchers can integrate multiple strategies and develop lithium-ion batteries or battery thermal management systems with enhanced safety performance. Keywords: Lithium-ion battery, Fire risks, Deep learning, Thermal management
DOI: 10.5281/zenodo.14213410
Language: English
Open Access status: green
OpenAlex ID: W6930714441
FWCI:
Cites:
Cited by:
Related to: