The Fault Diagnosis System of Electric Locomotive Based on MAS

The efficiency and the effect of fault diagnosis are increasingly important for electric locomotive. However, due to the complexity of the locomotive system configuration and the locomotive inferior running condition, it is much difficult to realize the fault diagnosis system on electric locomotive. In this paper, a multi-objective fault diagnosis algorithm based on multi-agent is proposed. It considers the correlation of different fault diagnosis and takes advantage of the capabilities of the multiple agents in communication and cooperation to estimate and recognize particular faults. Furthermore, a fault diagnosis system architecture is designed based on multi-agent system to support diagnosing fault online for train drivers. The practical application results show the efficiency of the proposed system.