GrC and SDG-Based Fault Diagnosis System and its Simulation Platform

Signed Directed Graph (SDG) model of hot nitric acid cooling system is constructed by using process experience method and system process, and the fault diagnosis rules are reasoned out by the reverse reasoning. The three states of nodes in SDG model are extended to seven states of nodes, which consider node deviating. Granular Computing (GrC) is introduced into SDG-based fault diagnosis, the redundant attributes in the system are simplified by heuristic reduction algorithm, and ultimately the optimal decision table is produced. The normal state of the system and equipment failure information is added to the decision table to ensure integrity of the decision table. In Matlab IDE, Graphical User Interface (GUI)-based fault diagnosis simulation platform achieves the solution of the above mentioned problem and proves the correctness and effectiveness of the proposed method.