Reliable fuzzy prognosability of decentralized fuzzy discrete-event systems and verification algorithm.

This paper investigates the problem of reliable fuzzy prognosis of decentralized fuzzy discrete event systems (FDESs) in face of possible unavailability of some local agents. Firstly, the notion of k-reliable fuzzy coprognosability is formalized for a decentralized FDES with m local agents, which requires that a k-reliable fuzzy coprognosable FDES can predict the fault to a certain degree in the manner of fuzziness even when m-k local agents are unavailable, where 1⩽k⩽m. Then, a reliable fuzzy copredictor is constructed to perform the reliable fuzzy prognosis of decentralized FDESs, and a necessary and sufficient condition of the k-reliable fuzzy coprognosability is presented. In particular, an algorithm based on reliable fuzzy copredictor is proposed to verify the k-reliable fuzzy coprognosability of FDESs.