基于贝叶斯网络的储油罐区事故情景分析  

Scenario analysis of oil storage tank farm accidents based on Bayesian network

作　　者：金辰浩 李铁[1] 关磊[2] 史伟静 刘亮 JIN Chenhao;LI Tie;GUAN Lei;SHI Weijing;LIU Liang(School of Civil and Resource Engineering,University of Science and Technology Beijing,Beijing 100083,China;China Academy of Safety Science and Technology,Beijing 100012,China;National Petroleum and Natural Gas Pipeline Network Group Co.,Ltd.,Beijing 100028,China)

机构地区：[1]北京科技大学土木与资源工程学院,北京100083 [2]中国安全生产科学研究院,北京100012 [3]国家石油天然气管网集团有限公司,北京100028

出　　处：《中国安全生产科学技术》2024年第8期120-128,共9页Journal of Safety Science and Technology

摘　　要：为确定储油罐区事故引发多米诺效应中各关键节点因素的重要性,基于情景构建过程中不同发展阶段的孕灾环境、处置措施和演化结果等要素,引入贝叶斯网络方法,提出1种在储油罐区事故情景构建下的演化网络,通过敏感性和影响强度分析,筛选出事故发展过程中的关键节点,并通过提炼事故支链来论证事故演化网络构建的合理性。研究结果表明:在事故进程中,火灾熄灭、火势隔离、人员撤离、危化品残存、全员撤离、组织搜救节点因素对多米诺效应的影响较大;提炼的各事故支链符合情景构建演化趋势,说明该事故演化网络具备一定合理性。研究结果可为储油罐区事故应急辅助决策技术的建设提供支持和参考。In order to determine the importance of various key nodes of the domino effect triggered by tank farm accidents,an evolution network under the scenario construction of oil storage tank farm accidents was proposed by introducing into the Bayesian network method based on the elements of disaster-inducing environment,disposal measures and evolution results at different development stages of scenario construction.Through the sensitivity and impact intensity analysis,the key nodes in the accident development process were screened out,and the accident branch chain was refined to demonstrate the reasonableness of the construction of accident evolution network.The results show that in the process of accident,the fire extinguishing,fire isolation,personnel evacuation,residual of hazardous chemicals,full evacuation,and organization of search and rescue nodes have a greater impact on the domino phenomenon,and each accident branch conforms to the evolution trend of scenario construction.The research results can provide support and reference for the construction of accident emergency assistant decision-making technology.

关 键 词：汽油储罐 池火灾 多米诺效应 情景构建 贝叶斯网络 

分 类 号：X937[环境科学与工程—安全科学]