Authors:
Felipe G. Cabral
;
Maria Z. M. Veras
and
Marcos V. Moreira
Affiliation:
Universidade Federal do Rio de Janeiro, Brazil
Keyword(s):
Failure Diagnosis, Modular Systems, Automata, Petri Nets.
Related
Ontology
Subjects/Areas/Topics:
Industrial Networks and Automation
;
Informatics in Control, Automation and Robotics
;
Robotics and Automation
;
Surveillance
Abstract:
In general, systems are formed by the composition of several modules, and may exhibit a large number of
states. The growth of the global system model with the number of components leads to a high computational
cost for diagnosis techniques. In order to circumvent this problem, in a recent work, a diagnosis scheme based
on the observation of the nonfailure behavior model of the system components, and their synchronization
due to observable events, is proposed. Although the computation of the global system model for diagnosis is
avoided, the estimated observed nonfailure language in this scheme can be a larger set than the actual observed
nonfailure language of the system, which leads to the notion of synchronous diagnosability. This scheme is
implemented using a diagnoser, called synchronized Petri net diagnoser (SPND). In this work, we propose the
addition of conditions for the observable transitions of the SPND, leading to a conditional synchronized Petri
net diagnoser (CSPND). We s
how that the addition of such conditions can cause a decrease in the observed
nonfailure language, and systems that are not synchronously diagnosable can be conditionally synchronously
diagnosable, and the delay bound can be smaller than using the synchronous diagnosis scheme.
(More)