Authors:
Mohamed Amine Atoui
;
Sylvain Verron
and
Abdessamad Kobi
Affiliation:
L’UNAM University, France
Keyword(s):
FDD, Data-Driven Methods, Model-based Methods, Incidence Matrix, Bayesian Networks, Water Heater System.
Related
Ontology
Subjects/Areas/Topics:
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Intelligent Fault Detection and Identification
;
Robotics and Automation
;
Surveillance
Abstract:
In this paper, we present an original FDD method. The interest of this method is her ability to coexist residuals and measures, under a same and a single tool. Indeed, our proposal is to combine two different Bayesian networks to FDD. A model-based method is associated to a data-driven method to enhance decision making on the system operating state. This method is evaluated on a simulation of a water heater system in some various circumstances.