Research On The Fault Diagnosis System For The Locomotive Based On Case-based Reasoning

With the rapid development of trail transportation in recent years, People’s living standards are improving unceasingly; therefore, more and more people choose to travel. As a maintenance department of railway transportation, Locomotive depot’s task is more, how to ensure the safety of locomotive running fast become priority for railway technology departments.At present, the locomotive depot has lead in several intelligent detection means in locomotive maintenance, which can’t adapt to the requirement of the railway modernization. Locomotive’s structure is complex and fault species is diversity. In recent years, it becomes more complicated and systematic, maintenance personnel’s experience and some simple equipment testing can’t satisfy the requirement of the locomotive modernization. Therefore, it is necessary to introduce intelligent diagnosis technology to locomotive fault diagnosis field.Case-based reasoning technology is a kind of artificial intelligence technology, which rises in recent years; it solves the current problem with historical case. The locomotive fault field has accumulated rich failure cases, so the using of case-based reasoning can greatly improve the locomotive fault diagnosis ability.A locomotive fault diagnosis system is designed on the basis of the characteristics of locomotive fault and case-based reasoning, which has the ability of fault diagnose and learning. The main research results are as follows:(1) The necessity of introduction of intelligent diagnosis in locomotive fault diagnosis is introduced. And the history of fault diagnosis and case-based reasoning are introduced too. The feasibility of using case-based reasoning in fault diagnosis is verified.(2) The concrete application of the case-based reasoning technology is expounded. The way that the case-based reasoning technology is applied to the locomotive fault diagnosis is studied. By combined the characteristics of locomotive fault and locomotive structure, a method of combining the object-oriented and database technology and using the database layer organization method in case representation is adopted. The rough set theory in attribute reduction is used; the neighbor retrieval methods in case similarity calculation are also used in the case of case retrieval. The model of multistage index is used to case retrieval.(3) The fault diagnosis system for the locomotive based on Case-based reasoning by Microsoft Visio Studio2010and Microsoft SQL Server2005is realized. Then the database design and system function module are introduced, and the function and realization of the main module are stated.The introduction of case-based reasoning can solve the problem of locomotive fault diagnosis and improve the efficiency of locomotives maintenance are demonstrated.