基于多源异构数据融合的化工安全风险动态量化评估方法

针对石化企业过程安全风险分析缺乏多源异构数据融合、难以解析风险动态时变机理的现状。首先基于改进的模糊Petri网建立风险动态传播模型，考虑初始事件失效、保护层失效和修正因子等参数，得到风险的实时变化概率；然后分析不同保护层失效对事故发生概率的影响，得到不同保护层的重要程度；最后，以正己烷缓冲罐溢流发生火灾爆炸事故为例进行动态风险计算，分析对比其在不同保护层失效下的事故发生概率变化，结果表明安全联锁系统对事故发生起关键作用，日常应加大对其检测力度，防止事故发生。

Dynamic quantitative assessment method of chemical safety risk based on multi-source heterogeneous data fusion

In view of the lack of multi-source heterogeneous data fusion in process safety risk analysis of petrochemical enterprises, it is difficult to analyze the dynamic time-varying mechanism of risk. First, it is established a dynamic risk propagation model based on the improved fuzzy Petri net, consider parameters such as initial event failure, protection layer failure, and correction factors to obtain the real-time change probability of risk. Then the influence of failure of different protective layers on the probability of accidents is analyzed and the importance degree of different protective layers is obtained. Finally, taking n-hexane buffer tank overflow fire and explosion accident as an example, the dynamic risk calculation was carried out to analyze and compare the change of accident probability under the failure of different protective layers. The results showed that the safety interlock system played a key role in the accident, and the daily detection should be strengthened to prevent the accident.