Research And Application Of The Resource Space Model In Mechanical Fault Diagnosis

With the rapid development of intelligent technology, people try to use knowledge engineering, expert system, artificial neural network and other technologies to simulate human experts' mechanical diagnosis methods, and integrate several experts' best knowledge experience, to implement mechanical failure analysis and diagnosis rapidly. However, the existing mechanical fault diagnosis systems are lack of semantic support, validity and completeness in knowledge base, which lead to the low degree of intelligent and accuracy in mechanical fault diagnosis systems. However, integration of exiting knowledge is an important way to improve the quantity and quality of knowledge base.The Resource Space Model (RSM), a kind of semantic data model, builds an appropriate multi-dimensional classification system for the resources of an application field, an organization or a person through classifying, storing, managing and locating. RSM manages the resources by using the normalized classification system, which can organize and govern large-scale heterogeneous resources effectively. So using RSM to realize the normalized management of mechanical fault diagnosis is demanded, This RSM for mechanical fault diagnosis knowledge, set up by applying RSM's theories and methods to the mechanical fault diagnosis system, can achieve the model's semantic features in the resources' organization and management.The main research content of this paper include: (1) apply RSM's theories to describe and classify common fault diagnosis knowledge of the rotating mechanism and build a diagnosis knowledge resource space model (RS); (2)use the radar software to realize the visual operation of rotating mechanism fault resource space model; (3) use the fuzzy resource space model to deal with the fuzzy query of rotating mechanism fault; according to the different query effects of resources' categories, build the fuzzy set, and transform the fuzzy query into calculating distant of resource and center point, analyze and improve the fuzzy resource query.