Aeroacoustic source term computation based on radial basis functions

Stefan Schoder*, Klaus Roppert, Michael Weitz, Clemens Junger, Manfred Kaltenbacher

*Korrespondierende/r Autor/-in für diese Arbeit

Publikation: Beitrag in einer FachzeitschriftArtikelBegutachtung

Abstract

In low Mach number aeroacoustics, the known disparity of length scales makes it possible to apply well‐suited simulation models using different meshes for flow and acoustics. The workflow of these hybrid methodologies include performing an unsteady flow simulation, computing the acoustic sources, and simulating the acoustic field. Therefore, hybrid methods seek for robust and flexible procedures, providing a conservative mesh to mesh interpolation of the sources while ensuring high computational efficiency. We propose a highly specialized radial basis function interpolation for the challenges during hybrid simulations. First, the computationally efficient local radial basis function interpolation in conjunction with a connectivity‐based neighbor search technique is presented. Second, we discuss the computation of spatial derivatives based on radial basis functions. These derivatives are computed in a local‐global approach, using a Gaussian kernel on local point stencils. Third, radial basis function interpolation and derivatives are used to compute complex aeroacoustic source terms. These ingredients are necessary to provide flexible source term calculations that robustly connect flow and acoustics. Finally, the capabilities of the presented approach are shown in a numerical experiment with a co‐rotating vortex pair.
Originalspracheenglisch
Seiten (von - bis)2051-2067
Seitenumfang17
FachzeitschriftInternational Journal for Numerical Methods in Engineering
Jahrgang121
Ausgabenummer9
DOIs
PublikationsstatusVeröffentlicht - 15 Mai 2020
Extern publiziertJa

ASJC Scopus subject areas

  • Ingenieurwesen (insg.)
  • Angewandte Mathematik
  • Numerische Mathematik

Fingerprint

Untersuchen Sie die Forschungsthemen von „Aeroacoustic source term computation based on radial basis functions“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren