A Study Of Fault Diagnosis Expert System Rule Attaining For Oil Pump Truck Based On Rough Set

With the significant development of modern industry and techniques, manufacturing facilities are growing sophisticated, large and smart, so it is not uncommon for machines to break down in our daily lives. It is greatly meaningful to perform early detection and then determine types and properties of breakdowns, in order to avoid devastating economic loss and massive casualties. This paper introduces the rough set theory into the analysis of the breakdown detection technique, aims at the bottleneck problem occurring when acquiring related knowledge of this method, sets YYBC-2aviation hydraulic power cart as the subject, to conduct deep-in-depth and detailed investigation.This paper describes a couple of methods for the breakdown detection technique, and does some analysis on status quo and future tendency as well. In addition, this paper summarizes the rough set theory and relevant sources, and then looks into the application of them. Based on the approach for attributes reduction on rough set theory, an advanced approach for attributes reduction on difference matrix is proposed. This methodology, in the light of the importance of attributes, employs heuristic algorithm to get results. According to samples, this approach improves the conventional method in terms of computational efficiency and steps.The traditional rough set reduction algorithms were identified many deficiencies. Based on these deficiencies, a new algorithm called value reduction algorithm based on the decision matrix was proposed. The algorithm introduces the idea of a heuristic algorithm. Based on the degree of importance of the attribute and make use of decision-making matrix to solve it. Compared with the pros and cons of the value reduction algorithm, this value reduction algorithm has a great advantage both on efficiency and computational complexity.This paper gives details of structures of the YYBC-2oil pump truck system. Based on analyzing the structural features of the pump car system, we can found fault diagnosis knowledge is characterized by a hierarchy. So the knowledge base is divided into two layers. The one-rate rule knowledge is got by utilized artificial acquisition, and the two-rate rule knowledge is got by RST acquisition. This paper describes the structure and the knowledge acquisition process of the fault diagnosis expert system in detail. Finally, it used oil pump truck system as a study object. We used rough set attribute reduction and value reduction algorithm to get the two-rate rule knowledge, so we improve the ability of the knowledge acquired.