基于贝叶斯网络的危化品爆炸事故电力系统风险评估模型*

为提高危化品爆炸事故电力应急预警的准确性，建立基于贝叶斯网络的危化品爆炸事故电力系统风险评估模型。基于危化品爆炸事故电力应急典型情景分析，建立综合考虑突发事件、承灾载体和应急管理等风险因素的贝叶斯网络结构。应用概率刻画风险因素信息的不确定性及其相互影响，定量分析事件后果。结合一般条件和典型情景等的应用实例，分析评价方法和风险因素对风险等级的影响。结果表明:该模型能够在危化品爆炸事故发生后，评价电力应急预警等级；能够在危化品爆炸事故发生前，分析典型情景的风险和风险因素的影响，为应急准备提供支持；“最大概率法”较“概率加权求和法”得出的事件等级可能较低。

Risk assessment model of electricity system in hazardous chemicals explosion accident based on Bayesian network

In order to improve the accuracy of electricity system’s emergency pre-warning for the hazardous chemicals accident,a risk assessment model of electricity system in the hazardous chemicals accident based on Bayesian network was established.Based on the analysis of typical power emergency scenarios in the hazardous chemicals accident,a Bayesian network structure comprehensively considering the emergency events,disaster carriers,emergency management and other risk factors was established.The probability was used to characterize the uncertainty of risk factors’information and their mutual influence,and the consequence of events were analyzed quantitatively.Combined with the application examples of general conditions and typical scenarios,the influence of assessment methods and risk factors on the risk grades was analyzed.The results showed that the model can assess the pre-warning level after an event occurs,and provide support for the early-warning.It can also analyze the risk of typical scenarios and the influence of risk factors before the event occurs,and provide support for the emergency preparedness.The event grade obtained by the“maximum probability method”maybe lower than that by the“probability-weighted summation method”.