Fault diagnosis model for coking system based on multi-agent

Aiming at the requirement on complexity, security and timeliness for fault diagnosis of coking process, the multi-agent fault diagnosis model for coking system was proposed by the hiberarchy analysis on coking system fault. Fuzzy relationship matrix knowledge model, learning mechanism based on genetic algorithm and fuzzy diagnosis reasoning process were discussed as an example of coking heating system fault diagnosis. It was validated through actual data that the model can diagnose fault efficiently and exactly and it can fulfill the requirement of production.