Title :
Bond graph UIO approach for fault components monitoring
Author :
Tarasov, E. ; Gahlouz, I. ; Sueur, C. ; Ould Bouamama, B.
Author_Institution :
LAGIS, Ecole Centrale de Lille, Villeneuve-d´Ascq, France
Abstract :
This paper presents a new method to design a sensor fault modeling and estimation using a bond graph approach with the unknown input observer (UIO) approach. Intrinsic solvability conditions are given with clear physical meanings due to the bond graph representation and a bond graph observer is proposed with an augmented model. With the augmented model, the sensor fault is modeled as an actuator fault, which justifies the UIO approach. An illustrative example which considers a real hydraulic system is included.
Keywords :
actuators; computability; fault diagnosis; observers; sensors; actuator fault; augmented model; bond graph UIO approach; bond graph observer; bond graph representation; fault components monitoring; intrinsic solvability conditions; sensor fault estimation design; sensor fault modeling design; unknown input observer approach; Biological system modeling; Calibration; Variable speed drives; Unknown Input Observer; bond graph; linear models; sensor fault modeling and estimation;
Conference_Titel :
Control and Fault-Tolerant Systems (SysTol), 2013 Conference on
Conference_Location :
Nice
DOI :
10.1109/SysTol.2013.6693855