主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2025, Vol. 44 ›› Issue (4): 439-447.

• • 上一篇    

基于N-K模型和贝叶斯网络的船舶载运汽车火灾风险耦合分析

王意如, 王彦富, 乔健, 李逊   

  1. （中国石油大学（华东） 机电工程学院，山东 青岛 266580）
  • 收稿日期:2024-06-05 修回日期:2024-08-03 出版日期:2025-04-15 发布日期:2025-04-15
  • 作者简介:王意如，中国石油大学（华东）硕士研究生，主要从事火灾风险评估方面的研究，山东省青岛市黄岛区长江西路 66号，266580。
  • 基金资助:
    国家自然科学基金面上项目（52171353）

Fire risk coupling analysis of ship carrying vehicles based on N-K model and Bayesian network

Wang Yiru, Wang Yanfu, Qiao Jian, Li Xun   

  1. (College of Mechanical and Electronic Engineering, China University of Petroleum (East China), Qingdao Shandong 266580, China)
  • Received:2024-06-05 Revised:2024-08-03 Online:2025-04-15 Published:2025-04-15

摘要： 风险之间由于传播效应和相互作用而产生的风险因素耦合，可能导致重大事故的发生。为了分析船舶载运汽车火灾事故风险的耦合性，提出N-K模型和贝叶斯网络（BN）模型相结合的一种新的定量分析方法。首先，通过对历史事故的统计分析，探讨船舶载运汽车火灾事故的原因，明确人员、机械、电气和环境因素的风险耦合类型；其次，构建N-K模型研究各因素之间的耦合关系和风险耦合机制；最后，根据上述结果构建BN模型并验证。此外，我们还将开展敏感性分析和反向推理分析等深入研究，以确保模型的准确性。结果表明：船舶载运汽车发生火灾概率最高的耦合类型是“人员-机械” 耦合，发生概率为14.795%；发生火灾概率最低的是“人员-机械-电气-环境”耦合，发生率为1.354%，但其风险耦合值最大，约为0.43。我们应当特别关注与人员操作和机械故障相关的风险因素，以预防这些因素的共存或多重叠加，进而有效降低船舶载运汽车火灾事故的发生，本研究的成果不仅对提升船舶载运汽车的防火规范具有指导意义，同时也为风险评估和消防救援工作提供了参考依据。

关键词: 船舶载运汽车火灾, 风险耦合分析, BN模型, N-K模型

Abstract: The coupling of risk factors between risks due to propagation effects and interdependence can lead to serious accidents. To analyze the fire risk coupling of ship carrying vehicles, this paper proposes a novel quantitative analysis method that integrates the N-K model with the Bayesian Network (BN) model. Firstly, through the statistical analysis of historical accidents, the fire causes of ship carrying vehicles accidents are investigated, and the types of human, mechanical, electrical, environmental factors and risk coupling are clarified. Second, the N-K model is constructed to study the coupling relationship between factors and the risk coupling mechanism. Finally, the BN is constructed and verified based on the above results. Sensitivity analysis and backward inference analysis are carried out. The results indicate that the coupling causing the highest probability of fire is “human-mechanical” with a probability of 14.795%; the coupling causing the lowest probability of fire is “human-mechanical-electrical-environmental” with a probability of 1.354%, yet it has the highest risk coupling value, approximately 0.43. It is of great importance to focus on the risk factors related to human and machinery to prevent the co-existence of the two or more, with the goal of reducing the occurrence of fire accidents. The results of this study will not only provide guidance for improving the fire safety of ship-carrying vehicles, but also provide an important reference for risk assessment and firefighting.

Key words: fire of ship carrying vehicles, risk coupling analysis, BN model, N-K model