基于多类别信息量的模糊Petri网故障推理策略  

Fault Reasoning Strategy Based on Multi-sort Information Value Under Fuzzy Petri Net

作　　者：秦文娟[1] 张骏[1] 宁宁[1] 

机构地区：[1]西北工业大学自动化学院,陕西西安710072

出　　处：《计算机测量与控制》2008年第7期949-951,共3页Computer Measurement &Control

摘　　要：针对模糊Petri网推理模型膨胀、推理过程繁琐的问题,研究基于信息论方法的推理过程简化策略;利用自信息和熵函数的概念,提出了一种多类别故障特征信息量准则,给出了多类别故障特征信息量的具体计算方法;制定系统的故障推理策略时,首先选择故障信息量最大的故障特征进行测试,然后依据故障特征的测试顺序建立模糊Petri网模型,最后根据模糊Petri网模型进行故障推理;结论证明,该方法能够有效地进行故障推理,实现系统故障的动态和快速搜索。As the reasoning model is growing bigger and the reasoning process is becoming more and more complicated in Fuzzy Petri Net model, a simplified strategy based on information theory is researched. Based on the conception of self-information and entropy, a new formula about multi--sort fault information is proposed, and also the way of how to calculate the information value is given. To frame the system reasoning strategy, the fault character which has the biggest information value is chosen to be tested first, then the Fuzzy Petri Net model is constructed according to the sequence of the test, finally the reasoning process is carried through by the Fuzzy Petri Net model. It is proved that, the new strategy can do fault reasoning effectively, and it can also realize the dynamic and fast search.

关 键 词：多类别 信息论 模糊PETRI网 故障诊断 

分 类 号：TH873.7[机械工程—仪器科学与技术]