Reference : Fault detection in mechanical systems based on subspace features
Scientific congresses and symposiums : Paper published in a journal
Engineering, computing & technology : Mechanical engineering
http://hdl.handle.net/2268/73212
Fault detection in mechanical systems based on subspace features
English
Nguyen, Viet Ha [Université de Liège - ULg > > > Doct. sc. ingé. (aérosp. & méca. - Bologne)]
Rutten, Christophe mailto [Université de Liège - ULg > Département d'aérospatiale et mécanique > LTAS - Vibrations et identification des structures >]
Golinval, Jean-Claude mailto [Université de Liège - ULg > Département d'aérospatiale et mécanique > LTAS - Vibrations et identification des structures >]
Sep-2010
International Conference on Noise and Vibration Engineering
No
No
International
Leuven
Belgium
International Conference on Noise and Vibration Engineering
21-23 september
KU Leuven
Leuven
Belgium
[en] KPCA ; NSA ; subspace ; condition monitoring ; statistics
[en] In the field of structural health monitoring or machine condition monitoring, the activation of nonlinear dynamic behavior complicates the procedure of damage or fault detection. Principal Component Analysis (PCA) is known as an efficient method for damage diagnosis. However, two drawbacks of PCA are the assumption of the linearity of the system and the need of many sensors. This article presents industrial applications of two possible extensions of PCA: Null subspace analysis (NSA) and Kernel PCA (KPCA). The advantages of NSA rely on its rapidity of use and its reliability. The KPCA method, through the use of nonlinear kernel functions, allows to introduce nonlinear dependences between variables. The objective is to address the problem of fault detection in mechanical systems using subspace-based methods. The detection is achieved by comparing the subspace features between the reference and a current state through statistics. Industrial data are used as illustration of the methods.
http://hdl.handle.net/2268/73212

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
Subspace_ISMA.pdfAuthor preprint375.04 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.