Abstraction-based failure diagnosis for discrete event systems

In this paper, we introduce the idea of abstraction-based diagnosability for large-scale composed discrete event systems that consist of multiple subsystems. To this end, we determine sufficient conditions such that diagnosability of the original system follows from diagnosability of an abstracted system model on a smaller state space. In addition, we prove that also the reverse implication is true if an additional requirement for the abstraction is fulfilled. Then, we show how our method can be applied to compute abstracted models for the diagnosability verification of composed systems without enumerating the whole system state space. In this way, considerable computational savings can be achieved as illustrated by a small manufacturing system example.