2018
DOI: 10.4208/jcm.1610-m2016-0494
|View full text |Cite
|
Sign up to set email alerts
|

Time Domain Boundary Element Methods for the Neumann Problem: Error Estimates and Acoustic Problems

Abstract: We investigate time domain boundary element methods for the wave equation in R 3 , with a view towards sound emission problems in computational acoustics. The Neumann problem is reduced to a time dependent integral equation for the hypersingular operator, and we present a priori and a posteriori error estimates for conforming Galerkin approximations in the more general case of a screen. Numerical experiments validate the convergence of our boundary element scheme and compare it with the numerical approximation… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
21
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
6
1

Relationship

5
2

Authors

Journals

citations
Cited by 18 publications
(24 citation statements)
references
References 14 publications
1
21
0
Order By: Relevance
“…Together with the a priori estimates for the time domain boundary element methods on screens [23,24], our results imply convergence rates for the p-version Galerkin approximations which are twice those observed for the quasi-uniform h-method in [22].…”
supporting
confidence: 58%
See 1 more Smart Citation
“…Together with the a priori estimates for the time domain boundary element methods on screens [23,24], our results imply convergence rates for the p-version Galerkin approximations which are twice those observed for the quasi-uniform h-method in [22].…”
supporting
confidence: 58%
“…, 5. The error remains stable over the time interval, reflecting the space-time variational discretization used [24]. A second right hand side investigates a plane-wave f 2 (t, (x, y, z) T ) = exp(−2/t 2 ) cos(ωt − k(x, y, z) T ) , with k = (3, 0.5, 0.1) and ω = |k|.…”
Section: Wave Equation Outside An Icosahedronmentioning
confidence: 99%
“…Compute right hand side f n−1 − f n − n−1 m=1 V n−m ψ m Solve system of linear equations (34) Store solution ψ n end for see [15] for details. We use ansatz functions in V 1,1 h,∆t .…”
Section: Algorithmic Considerationsmentioning
confidence: 99%
“…Let ε > 0. a) Let φ be the solution to the hypersingular integral equation W φ = g and φ β h,∆t the best approximation in the norm of H r σ (R + , H Note that the energy norm associated to the weak form of the single layer integral equation (7) is weaker than the norm of H 1 σ (R + , H − 1 2 (Γ)) and stronger than the norm of H 0 σ (R + , H − 1 2 (Γ)), according to the coercivity and continuity properties of V on screens [13]. Similarly, for the weak form of the hypersingular integral equation (10), the energy norm is weaker than the norm of H 1 σ (R + , H Remark C. Together with the a priori estimates for the time domain boundary element methods on screens [13,15], Corollary B implies convergence rates for the Galerkin approximations, which recover those for smooth solutions (up to an arbitrarily small ε > 0) provided the grading parameter β is chosen sufficiently large.…”
Section: Introductionmentioning
confidence: 99%
“…. , N t (see [24] for further details). From the convolution structure in time, we obtain: For arbitrary n ∈ {1, .…”
Section: 1 Marching-on-in-time Scheme For the Variational Equalitymentioning
confidence: 99%