故障Petri网的概率变迁方法研究  

Research on probability transition method for fault Petri net

作　　者：盛晟[1] 肖明清[1] 赵亮亮[1] 文莹[1] 胡斌[1] 

机构地区：[1]空军工程大学自动测试系统实验室,西安710038

出　　处：《仪器仪表学报》2014年第3期714-720,共7页Chinese Journal of Scientific Instrument

摘　　要：为了对故障建模中的不确定性问题进行描述,并降低故障推理的计算复杂度,定义了故障Petri网的概率变迁方法,对概率变迁及其激发规则作了有别于传统Petri网的定义;将系统故障解释为源库所故障或者变迁的激发功能故障;依据变迁及其外延的一般形式,给出了故障Petri网的概率变迁分析方法,由此提出了一种故障诊断算法。在实例中,设计实验对所提出的故障诊断算法进行验证,结果表明算法能够快速、有效、可靠地诊断出系统故障。In order to describe the uncertainty in fault modeling and reduce the computing complexity of fault reasoning,this paper gives the definition of probability transition method for Fault Petri net.The concepts of probability transition and its fire rule are different from those in the definition of traditional Petri net.The system fault is explained to contain the fault of source place and the fault of the fire function of transition.The probability transition analysis method for fault Petri net is given according to the study on the general form of transition and its extension,and based on which,a fault diagnosis algorithm is given.Taking a power system as the example,the fault diagnosis algorithm was verified with the designed experiment.The results show that the proposed algorithm can diagnose the system faults fast,effectively and reliably.

关 键 词：故障 PETRI网 概率变迁 诊断 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置] TP306.3[自动化与计算机技术—控制科学与工程]