Scholarlink

<p>Lithium battery fires can lead to severe casualties and significant property losses. Proactively evaluating and predicting lithium battery hazards enables timely preventive measures, thereby mitigating the severity of potential fire incidents through enhanced safety management. Therefore, conducting risk assessments and implementing safety measures for lithium battery fires is essential. Firstly, this study classified lithium battery fire risks into four grades and summarized the key factors contributing to battery fires. Secondly, it developed an integrated Fuzzy Bayesian Network (FBN) model that innovatively combines fishbone diagram analysis with probabilistic reasoning to quantitatively assess the complex interdependencies among fire-inducing factors. To validate the model’s effectiveness, a case study was conducted, and the results align with the actual risk levels of historical accidents, confirming the model’s reliability. This model can serve as a valuable reference for lithium battery fire risk evaluation. Furthermore, based on the case study findings and the constructed risk assessment model, a corresponding safety management system was established. This system helps reduce accident occurrence and minimize the severity of their consequences.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Risk Assessment Model and Safety Management of Lithium Battery Fire Based on Bayesian Network

  • 期刊论文

  • 发表日期：

    2026年2月6日

查看全文
  • Meijuan Lan,
  • Binshan Yu,
  • Jianpeng Zeng,
  • Yujie Huang

摘要

Lithium battery fires can lead to severe casualties and significant property losses. Proactively evaluating and predicting lithium battery hazards enables timely preventive measures, thereby mitigating the severity of potential fire incidents through enhanced safety management. Therefore, conducting risk assessments and implementing safety measures for lithium battery fires is essential. Firstly, this study classified lithium battery fire risks into four grades and summarized the key factors contributing to battery fires. Secondly, it developed an integrated Fuzzy Bayesian Network (FBN) model that innovatively combines fishbone diagram analysis with probabilistic reasoning to quantitatively assess the complex interdependencies among fire-inducing factors. To validate the model’s effectiveness, a case study was conducted, and the results align with the actual risk levels of historical accidents, confirming the model’s reliability. This model can serve as a valuable reference for lithium battery fire risk evaluation. Furthermore, based on the case study findings and the constructed risk assessment model, a corresponding safety management system was established. This system helps reduce accident occurrence and minimize the severity of their consequences.