Fault diagnosis in DESs modeled by partially observed Petri nets

Abstract

In this paper, we focus on fault diagnosis in discrete event systems (DESs) which are modeled by partially observed Petri nets. We consider not only the case where faults occur either on transitions or places, but also a more general case where faults occur on both transitions and places at the same time. Some faults cannot be diagnosed directly due to the unobservability of some transitions and places in a partially observed Petri net. We propose an approach to diagnose the faults that cannot be diagnosed directly using by an algebraic decoding technique. More specifically, we employ Nearest Neighbour Decoding (NND) to determine event occurrences in a Petri net based on the observations from sensors in places and transitions, and then the expected marking can be calculated. Fault diagnosis is based on the difference between the expected marking and the observed marking.