故障诊断专家系统的新型知识网络推理模型  

Areasoning model based on a new knowledge network for fault diagnosis expert system

作　　者：黄波[1] 倪重匡 

机构地区：[1]复旦大学计算机科学系,上海200433

出　　处：《复旦学报（自然科学版）》1997年第6期625-632,共8页Journal of Fudan University：Natural Science

摘　　要：介绍故障诊断专家系统中的一种新型知识网络表示法，并给出了详细的推理算法与故障判决算法．此知识网络的推理通过消息传递来实现，具有分布式特性，且在整个知识网络的推理与故障判决中融入了不精确推理算法，故能适合大多数的故障诊断专家系统模型．In this paper, a new representation for the knowledge network of fault diagnosis expert system is introduced. The detailed reasoning algorithm and fault-finding algorithm are also presented.Through the message propagation among the nodes, this kind of knowledge network can also reason parallelly, which makes it has the characteristic of a distributed system. lt has the ability to infere with uncertainty. This reasoning model can be used to build most of fault diagnosis expert systems.

关 键 词：故障诊断 专家系统 知识网络 不精确推理 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]