Optimistic Fault Diagnosis in Discrete Event Systems by Labeled Petri Nets and Basis Markings

This paper deals with the fault diagnosis problem of discrete event systems modeled with labeled Petri nets. Its main contributions are threefold. First, depending on whether a diagnosis function examines the fault transitions that possibly occur after the last observed event, we formally divide the diagnosis functions into two types: optimistic and pessimistic, which aims to facilitate the exploration of different diagnosis approaches. Second, a framework is proposed, which extends a given diagnosis approach for Petri nets to the case of labeled Petri nets. The main idea of the framework is to compute and combine the diagnosis results of observable transition sequences corresponding to an observed word. Third, we convert a basis-marking-based approach that is originally pessimistic to the optimistic case and prove the correctness of this conversion.