2012
DOI: 10.1007/s00466-012-0702-1
|View full text |Cite
|
Sign up to set email alerts
|

Parametric finite elements, exact sequences and perfectly matched layers

Abstract: The paper establishes a relation between exact sequences, parametric finite elements, and perfectly matched layer (PML) techniques. We illuminate the analogy between the Piola-like maps used to define parametric H 1 -, H(curl)-, H(div)-, and L 2 -conforming elements, and the corresponding PML complex coordinates stretching for the same energy spaces. We deliver a method for obtaining PML-stretched bilinear forms (constituting the new weak formulation for the original problem with PML absorbing boundary layers)… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
21
0

Year Published

2013
2013
2020
2020

Publication Types

Select...
6
1
1

Relationship

1
7

Authors

Journals

citations
Cited by 34 publications
(23 citation statements)
references
References 19 publications
(18 reference statements)
2
21
0
Order By: Relevance
“…The method also incorporates a Perfectly Matched Layer (PML) for truncation of the computational domain, as first proposed by Bérenger (1994). Our understanding and implementation of the PML is consistent with the one provided by Chew and Weedon (1994), which is based on a change of coordinates of the governing equations into the complex plane within the PML layer (Matuszyk et al 2012; Matuszyk and Demkowicz 2012). The coordinates are transformed according to the formula where x j and X j represent unstretched and stretched coordinates, respectively, k = 2π/λ min (λ min is the shortest anticipated wavelength for a given frequency and all the material properties) and The coordinates and define the PML region for the coordinate x j and δ j denotes its thickness.…”
Section: Methodssupporting
confidence: 60%
“…The method also incorporates a Perfectly Matched Layer (PML) for truncation of the computational domain, as first proposed by Bérenger (1994). Our understanding and implementation of the PML is consistent with the one provided by Chew and Weedon (1994), which is based on a change of coordinates of the governing equations into the complex plane within the PML layer (Matuszyk et al 2012; Matuszyk and Demkowicz 2012). The coordinates are transformed according to the formula where x j and X j represent unstretched and stretched coordinates, respectively, k = 2π/λ min (λ min is the shortest anticipated wavelength for a given frequency and all the material properties) and The coordinates and define the PML region for the coordinate x j and δ j denotes its thickness.…”
Section: Methodssupporting
confidence: 60%
“…However, our methods can very easily be applied to any other conceivable variational formulation. Complex coordinate stretching has been discussed in an earlier work for different equations using Bubnov-Galerkin methods [74]. Inspired by this and to elaborate the differences, we present two possible categories of PML-consistent ultraweak variational formulations.…”
Section: Introductionmentioning
confidence: 89%
“…(2) Alternatively, first multiply with test functions and integrate by parts to construct a variational formulation in the stretched coordinates, then pull back the corresponding stretched variational formulation to spatial coordinates. In classical functional settings, it is well-known that these two approaches are identical in that they return the same ultimate variational formulation [74]. However, there are subtle differences to acknowledge in alternative functional settings.…”
Section: Acousticsmentioning
confidence: 99%
“…The formulation also includes a PML to impose radiation (Sommerfeld) conditions, thereby enabling a finite truncation of the computational domain. We introduce the PML into the weak form given by equation 1 and following the method developed by Matuszyk and Demkowicz (2013), namely, by introduction of the stretching JacobianJ ¼ detðJÞ and matrixà ¼JJ −1J−T , where the complex coordinate stretching JacobianJ and its determinant are given bỹ…”
Section: Formulation and Numerical Methodsmentioning
confidence: 99%