基于动态贝叶斯网络的储罐区多米诺效应事故研究  

Research on Domino Effect Accidents in Tank Farm Based on Dynamic Bayesian Networks

作　　者：王璇 宫博 WANG Xuan;GONG Bo(School of Environmental and Safety Engineering,Shenyang University of Chemical Technology,Shenyang Liaoning 110142,China)

机构地区：[1]沈阳化工大学环境与安全工程学院,辽宁沈阳110142

出　　处：《当代化工》2024年第11期2755-2759,共5页Contemporary Chemical Industry

基　　金：辽宁省教育厅科学研究面上项目(项目编号:LJKZ0440)。

摘　　要：化工储罐区存放的多为易燃、易爆的危险化学品,若储存不当发生泄漏,极易引发多米诺效应,造成财产损失甚至是更为灾难性的后果。选取某化工企业储罐区,分析多米诺效应事故扩展过程中的关键时间节点,运用Probit模型和贝叶斯方法计算储罐多米诺效应扩展概率;利用动态贝叶斯网络方法和GeNie软件,确定储罐区多米诺效应事故的动态演变过程。结果表明:T2、T3、T4、T5、T6、T7、T8储罐在0~10 min、>10~13 min、>13~18 min、>18~23 min和>23 min 5个时间段内多米诺事故扩展概率分别为0.930、0.930、0.607、0.526、0.517、0.597、0.584。该结果为后续化工储罐区风险评估提供理论依据,并为多米诺效应事故应急决策提供一定的参考。The chemical storage tank area stores mostly flammable and explosive hazardous chemicals.If it leaks due to improper storage,it will easily lead to domino effect,causing property damage and even more catastrophic consequences.Selecting the storage tank area of a certain chemical enterprise as research object,the key time nodes during the expansion process of the domino effect accident were analyzed,and Probit model and Bayesian method were used to calculate the probability of domino effect expansion of the storage tank;using the dynamic Bayesian network method and GeNie software,the dynamic evolution process of the domino effect accident in the storage tank area was determined.The results showed that the probabilities of domino accident expansion for T2,T3,T4,T5,T6,T7,and T8 storage tanks during the five time periods of 0~10 min,>10~13 min,>13~18 min,>18~23 min,and>23 min were 0.930,0.930,0.607,0.526,0.517,0.597,and 0.584,respectively.The results provide a theoretical basis for subsequent risk assessment of chemical storage tank areas and a certain reference for emergency decision-making of domino effect accidents.

关 键 词：时间序列 动态贝叶斯网络 多米诺效应 储罐区 

分 类 号：TE88[石油与天然气工程—油气储运工程]