Process Fault Diagnostics using the Integrated Graph Model

Abstract

This paper reveals some experiences gained during the development of a knowledge-based expert system for on-line fault detection and diagnosis. The symptom tree model and fault-consequence digraph developed in our previous research showed that they can be applied effectively in real-time fault diagnosis of plants, particularly for large and complex ones. In the current study, the two knowledge representation models, the symptom tree and the fault-consequence digraph, is synthesized. The symptom tree can be synthesized using mini-fault tree concept. The fault-consequence digraph can be derived using qualitative simulation. The synthesis can be performed using the material balance, energy balance, momentum balance and the equipment constraints. This new model will be tested and verified in the real-time application on a BTX process or a crude unit process. The reliability and flexibility are expected to be greatly enhanced by the results of the test conducted on a real plant. This automatic generation gives the generality, reliability, effectiveness and usefulness in building an expert system for process diagnosis. Nexpert Object for the expert system shell and SUN4 workstation for the hardware platform are used. TCP/IP for a communication protocol and interfacing to a dynamic simulator, SPEEDUP, for a dynamic data generation are also studied.