Complex Network Analysis on Fire Risk Evolution of Hazardous Chemical Tanks

作者：Jing-jing Li, Xiao-ying Xu, Xuanya Liu · 年份：2018 · DOI：10.2991/rac-18.2018.64 · 研究领域：Environmental Science

In order to control fire chain risk of hazardous chemical tanks, a method of risk analysis and control on fire evolution of hazardous chemicals is proposed based on the complex networks theory. Firstly, fire risk factors of hazardous chemical tanks were analyzed and screen initial risk events, transitive risk events and terminate risk events. A network model of risk evolution was made based on the logical relationship of risk events. Then the out-in degrees and the clustering coefficients were discussed for risk analysis and the key events of risk evolution were presented. Finally, considering the probability of events occurrence and transformation, the model was converted into a directed network with weight of events. And the shortest paths of risk evolution were calculated by using the Dijkstra algorithm. The risk control measures on disconnecting chains were presented. The results showed that the clustering coefficient of the model was 0.0424, the network aggregation degree was relativelylow.The weights of all the shortest paths were no more than 15. The research can provide the theoretical foundation for the inhibition of risk transfer and control of accident evolution, which has practical significance for guarding the continuous safety operation. key words: fire accidents; risk evolution; complex network; disconnect chains 摘要 为控制危险化学品储罐火灾连锁风险, 基于复杂网络理论提出一种针对储罐火灾 风险演化的分析与控制思路。辨识危险化学 品罐区火灾风险因素,筛选初始风险事件、 风险传递事件和末端风险事件,将风险事件 之间的逻辑关系转化为风险演化网络结构 模型。通过网络节点的出入度分析、聚类系 数分析,确定风险演化的...