基于BN的锅炉超压爆炸事故原因推理与故障诊断

分析了锅炉爆炸的原因,讨论了事故树与贝叶斯网络的特点,编制了以"锅炉超压爆炸事故"为顶事件的事故树,将其转化为锅炉超压爆炸事故贝叶斯网络拓扑结构,利用贝叶斯网络的技术优势,针对锅炉超压爆炸事故贝叶斯网络拓扑结构分别进行了原因推理与诊断推理。实例应用表明,在原因推理中,根节点V_1(压力上升)发生时锅炉超压爆炸事故发生的可能性最高;在诊断推理中,当锅炉超压爆炸事故发生时,根节点V_1(压力上升)一定发生,其次为V_4(定压不准),V_(11)(未装疏水管)发生的可能性最小,为企业进行锅炉超压爆炸事故预测与故障诊断提供了理论依据。

Analyses on Boiler Overpressure Explosion Causes and Fault Diagnosis Based on BN

In this paper ,the causes of boiler explosion are analyzed ,the features of accident tree and the Bayesian network are discussed ,the constructed accident tree of the boiler overpressure explosion transformed into Bayesian network topology of boiler overpressure accident and the causal reasoning and diagnostic reasoning of Bayesian network topology of the boiler overpressure explosion are analyzed by the Bayesian network technology .The examples show that the possibility of the boiler overpressure explosion is highest when the root node V1 (rising pressure ) occurs for causal reasoning .In the diagnosis rea-soning ,when the oiler overpressure explosion occurs ,the root node V1 (rising pressure ) must be occurred ,the following is V4 (inaccurate pressure ) and the minimum probability is the root node V11 (not installing drain pipe ) ,which provides theo-retical basis for the boiler overpressure accident prediction and fault diagnosis of enterprises .