A wave-based optimization approach of curved joints for improved defect detection in waveguide assemblies - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Sound and Vibration Année : 2020

A wave-based optimization approach of curved joints for improved defect detection in waveguide assemblies

Résumé

A wave-based numerical approach is proposed for the detection of defects in waveguide assemblies with curved joints. Within this framework, the wave finite element (WFE) method is used. It provides an efficient numerical means for computing waves in one-dimensional periodic structures (waveguides), and assessing the reflection and transmission coefficients of waves around defects and curved joints. A so-called apparent reflection matrix of the defects, which takes into account the influence of the joints on the reflected signals recorded at some measurement point at the beginning of a waveguide assembly, is proposed. This appears to be the relevant criterion for detecting defects. As it turns out, an optimization procedure for the design of curved joints can be proposed to magnify the amplitude of the reflected signals issued from defects. Numerical experiments are carried out on 2D waveguide assemblies, with one or two curved joints which are parameterized with respect to their radius and angle of curvature. Optimized values of these parameters can be found which magnify the reflected signals issued from several kinds of defects. Time response simulations are finally undertaken to highlight the relevance of the proposed approach.
Fichier principal
Vignette du fichier
manuscript.pdf (1.51 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02442812 , version 1 (16-01-2020)

Identifiants

Citer

Vivien Denis, Jean-Mathieu Mencik. A wave-based optimization approach of curved joints for improved defect detection in waveguide assemblies. Journal of Sound and Vibration, 2020, 465, pp.115003. ⟨10.1016/j.jsv.2019.115003⟩. ⟨hal-02442812⟩
106 Consultations
83 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More