A Fault Diagnosis Method Based on Simulation of Real Conditions

This paper proposed a new method of fault diagnosis for instrumentation and control system based on the simulation of real conditions. The first step was to establish a simulation system, in which the logic was consistent with that of the real system. Then signals in the real conditions were sampled as the source of the simulation system and executed in the operator of the simulation system. It utilized Eigen value comparison and wavelet transform to evaluate the deviation between the signal in real system and that in simulation system. The fault was detected based on the comparison between the derived deviation and the threshold by setting or experience. The efficacy of the methodology was demonstrated by implementation and validation in the field.