Study on Electric Fault Diagnosis Model Based on MAS

Under the new situation that the railway transportation organization pattern is gradually transforming and extending to the direction of centralized dispatching,comprehensive systems integration,intelligent management and information sharing and integrating,in view of the fact that the functions of signal monitoring systems of high-speed railways still have defects,the distributed artificial intelligence technology was introduced into the fault diagnosis system of signal devices.Utilizing the strong MAS ability to solve complicated system problems,the MAS-based fault diagnosis system was established.By indication of facing the Agent knowledge,the center BDI models were constructed,which consisted of central modules of data acquisition Agent,diagnosis Agent,event analysis Agent and management Agent.The hybrid Agent structure was extended.The conventional diagnosis system without self-learning ability was designed into the low-coupling high-cohesion diagnosis system with self-learning ability and parallel to the MAS diagnosis system.The proposed system has good reliability,augmentabilily and robustness so that the level of fault diagnosis and monitoring management is raized.