Stable Perfectly Matched Layers with Lorentz transformation for the convected Helmholtz equation - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Article Dans Une Revue Journal of Computational Physics Année : 2021

Stable Perfectly Matched Layers with Lorentz transformation for the convected Helmholtz equation

Résumé

Perfectly Matched Layers (PMLs) appear as a popular alternative to non-reflecting boundary conditions for wave-type problems. The core idea is to extend the computational domain by a fictitious layer with specific absorption properties such that the wave amplitude decays significantly and does not produce back reflections. In the context of convected acoustics, it is well-known that PMLs are exposed to stability issues in the frequency and time domain. It is caused by a mismatch between the phase velocity on which the PML acts, and the group velocity which carries the energy of the wave. The objective of this study is to take advantage of the Lorentz transformation in order to design stable perfectly matched layers for generally shaped convex domains in a uniform mean flow of arbitrary orientation. We aim at presenting a pedagogical approach to tackle the stability issue. The robustness of the approach is also demonstrated through several two-dimensional high-order finite element simulations of increasing complexity.
Fichier principal
Vignette du fichier
Stable_Perfectly_Matched_Layers_with_Lorentz_transformation_for_the_convected_Helmholtz_equation_Post.pdf (4.06 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02556182 , version 1 (27-04-2020)
hal-02556182 , version 2 (10-02-2021)

Identifiants

Citer

Philippe Marchner, Hadrien Beriot, Xavier Antoine, Christophe Geuzaine. Stable Perfectly Matched Layers with Lorentz transformation for the convected Helmholtz equation. Journal of Computational Physics, 2021, ⟨10.1016/j.jcp.2021.110180⟩. ⟨hal-02556182v2⟩
245 Consultations
204 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More