The necessity for the diagnosis of rotating machinery, which is widely used in the industry, is gradually increasing. A great deal of research has been done in the field of automatic diagnosis of the rotating machinery using computers instead of skilled operators. In this paper, an expert system called VIBEX has been developed to aid plant operators in diagnosing the causes of abnormal vibration in rotating machinery. In order to automate the diagnosis of the rotating machinery, a decision table based on the Sohre's cause-symptom matrix is used as a probabilistic method for diagnosing 42 causes of abnormal vibration. Also a decision tree, which is used as the acquisition of structured knowledge in the form of concepts, is introduced to build a knowledge base which is indispensable in expert systems. The decision tree is a technology used to build knowledge-based systems using inductive inferences from examples, and it also plays an important role as a vibration diagnostics tool. The system has been successfully implemented in personal computer environment and written in C++. To validate system performance, the diagnostic system has been tested with an example using two diagnostic methods.
展开▼
