A Method for Representation of Knowledge and Inference Based on MAS in Fault Diagnosis System

Abstract

With the development of the distributed artificial intelligence system, multi-agent system (MAS) has been applied in construction of large-scale fault diagnosis systems. Procedure of construction of the knowledge base about fault diagnosis in conventional knowledge models cannot satisfy demands of a synchronism and concurrency of the system. To solve problems mentioned above, a new WFPN model and the corresponding fuzzy reasoning algorithm are proposed in this paper. The effectiveness of this method is verified by simulation. Results show that this model has advantage in building of large-scale fuzzy fault diagnosis systems over conventional knowledge models.