Sparse Representations of Acoustic Emission Signal:Application to Damage Analysis in Glass Fiber Reinforced Composites

Type : Article de conférence
Auteur(s) :  Abida Satour, Silvio MONTRESOR, Mourad Bentahar, Rachid El GUERJOUMA, Fouad Boubenider
Année :  2013
Domaine : Sciences des matériaux
Conférence: 13th International Symposium on Nondestructive Characterization of Materials (NDCM-XIII)
Lieu de la conférence: 
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF :  (.pdf)
Mots clés :  acoustic emission, Signal processing, Ultrasonic, precursor damage

Résumé : 

Sparse representation of signals has been subject of many advances during recent years especially in the field ofapplied physics. The problem is to search for the most compact representation of signals in terms of linearcombinations of elementary waveforms in an over-complete dictionary. The underlying hypothesis in most ofsparse representation problems is that if one can associate a sparse representation to a given signal, its structurereveals information directly linked to the physical nature of the addressed problem. From this point of view it hasbeen shown that sparse representation problems have closed connexions with inverse problems and sourceseparation problems. In this contribution, an application of sparse representation techniques to identify damagemechanisms from Acoustic Emission (AE) signals is presented. The sparse representation is which is based onthe continues wavelet transform. All experiments concern investigations realized on glass fibre reinforcedpolymer composites (GFRP).