Application And Research Of Using MBD In Fault Daignosis For Traction Substation

To aid the operators in finding the reasons quickly and correctly when the power supplying system of electrify railway occured fault, a lot of artificial intelligence systems, especially expert systems, for the analysis and diagnosis of fault were studied and developed in last decades. Because the fault diagnosis expert systems were based on experts' experiences which need a long time to obtain, the period for developing an expert system was very long, and likely couldn't catch up the step of equipments' updating. To avoid the difficulties in collecting experts' experiences, the application and research of using Model-Based Diagnosis (MBD) in traction substation's fault diagnosis was done in this paper. In MBD, the princirple and the fault models of traction substation were used for its fault diagnosis.Firstly, the basic ideas, the basic processes, the common classification and the common problems of MBD were briefly introduced. Then, to the problem of searching minimal conflicts, the concept and the theory of potential minimal conflicts were introduced, while the searching method was studied. A potential minimal conflicts searching algorithm named Relation Guiding Algorithm (RGA) was proposed in this paper. In this Algorithm, an initial environment was produced by using the informations implicated in the the system's analytical redundancies, then a larger environment that enclosing the potential minimal conflicts was gotten by properly enlargeing the initial environment. Through using dimidiate eliminating method, all the base elements of the potential minimal conflicts were sorted out, which means that the potential minimal conflicts can be found out. A test by comparing with two usual searching algorithms was done, and it indicated that RGA can quickly find out all potential minimal conflicts of the system.Secondly, with the considering of the trait of traction substation's structure and the demands in traction substation's fault diagnosis, a scheme about using the mixed reasoning MBD in traction substation's fault diagnosis was proposed, and its modeling rules for constructing elements and system were established. The searching methods and the diagnosis's discriminating strategies involved in this scheme were discussed.In the end, an example in which an AT traction substation in Hefi-Nanning electrify railway was used as the diagnosing object was given. In this example, the uniform models were constructed and the system description of the AT traction substation was given. In addition, the diagnosis experiment was done with a set of observations of the AT traction substation. According to the result of this experiment, it was verified that the scheme proposed in this paper was feasible and available.