The optimised Schwarz method and the two-Lagrange multiplier method for heterogeneous problems in general domains with two general subdomains

David Neil Greer, Sébastien Loisel

Research output: Contribution to journalArticle

Abstract

The optimised Schwarz method and the related two-Lagrange multiplier (2LM) method are nonoverlapping domain decomposition methods which can be used to numerically solve boundary value problems. Local Robin problems are solved on each subdomain in parallel to approximate the global solution, where a careful choice of Robin parameters leads to faster convergence. The 2LM method involves the solution of a nonsymmetric linear system that is usually solved with a Krylov subspace method such as GMRES. The speed of convergence of GMRES can be estimated using a conformal map from the exterior of the field of values of the system matrix to the interior of the unit disc. In this article we consider an elliptic PDE problem with a jump in diffusion coefficients across the interface between the subdomains. We approximate the field of values of the 2LM system matrix by a rectangle, R, in ℂ and provide optimised Robin parameters that ensure R is “well conditioned” in the sense that GMRES converges quickly. We derive convergence estimates for GMRES and consider the behaviour asymptotically as the mesh size h becomes small and the jump in coefficients becomes large. We observe, for our choice of Robin parameters, that increasing the jump in coefficients increases the convergence rate of GMRES. Numerical experiments are performed to verify the theoretical results.

Original languageEnglish
Pages (from-to)737-762
Number of pages26
JournalNumerical Algorithms
Volume69
Issue number4
Early online date30 Sep 2014
DOIs
Publication statusPublished - Aug 2015

Fingerprint

Schwarz Methods
Lagrange multiplier Method
GMRES
Field of Values
Jump
Robin Problem
Nonsymmetric Linear Systems
Nonoverlapping Domain Decomposition
Conformal Map
Convergence Estimates
Krylov Subspace Methods
Elliptic PDE
Speed of Convergence
Domain Decomposition Method
Coefficient
Global Solution
Rectangle
Diffusion Coefficient
Unit Disk
Rate of Convergence

Keywords

  • Domain decomposition
  • GMRES convergence
  • Heterogeneous media
  • Optimised schwarz method
  • Two-lagrange multiplier method

Cite this

@article{dca56b04471c473e992945d80abe5dd5,
title = "The optimised Schwarz method and the two-Lagrange multiplier method for heterogeneous problems in general domains with two general subdomains",
abstract = "The optimised Schwarz method and the related two-Lagrange multiplier (2LM) method are nonoverlapping domain decomposition methods which can be used to numerically solve boundary value problems. Local Robin problems are solved on each subdomain in parallel to approximate the global solution, where a careful choice of Robin parameters leads to faster convergence. The 2LM method involves the solution of a nonsymmetric linear system that is usually solved with a Krylov subspace method such as GMRES. The speed of convergence of GMRES can be estimated using a conformal map from the exterior of the field of values of the system matrix to the interior of the unit disc. In this article we consider an elliptic PDE problem with a jump in diffusion coefficients across the interface between the subdomains. We approximate the field of values of the 2LM system matrix by a rectangle, R, in ℂ and provide optimised Robin parameters that ensure R is “well conditioned” in the sense that GMRES converges quickly. We derive convergence estimates for GMRES and consider the behaviour asymptotically as the mesh size h becomes small and the jump in coefficients becomes large. We observe, for our choice of Robin parameters, that increasing the jump in coefficients increases the convergence rate of GMRES. Numerical experiments are performed to verify the theoretical results.",
keywords = "Domain decomposition, GMRES convergence, Heterogeneous media, Optimised schwarz method, Two-lagrange multiplier method",
author = "Greer, {David Neil} and S{\'e}bastien Loisel",
year = "2015",
month = "8",
doi = "10.1007/s11075-014-9923-z",
language = "English",
volume = "69",
pages = "737--762",
journal = "Numerical Algorithms",
issn = "1017-1398",
publisher = "Springer",
number = "4",

}

TY - JOUR

T1 - The optimised Schwarz method and the two-Lagrange multiplier method for heterogeneous problems in general domains with two general subdomains

AU - Greer, David Neil

AU - Loisel, Sébastien

PY - 2015/8

Y1 - 2015/8

N2 - The optimised Schwarz method and the related two-Lagrange multiplier (2LM) method are nonoverlapping domain decomposition methods which can be used to numerically solve boundary value problems. Local Robin problems are solved on each subdomain in parallel to approximate the global solution, where a careful choice of Robin parameters leads to faster convergence. The 2LM method involves the solution of a nonsymmetric linear system that is usually solved with a Krylov subspace method such as GMRES. The speed of convergence of GMRES can be estimated using a conformal map from the exterior of the field of values of the system matrix to the interior of the unit disc. In this article we consider an elliptic PDE problem with a jump in diffusion coefficients across the interface between the subdomains. We approximate the field of values of the 2LM system matrix by a rectangle, R, in ℂ and provide optimised Robin parameters that ensure R is “well conditioned” in the sense that GMRES converges quickly. We derive convergence estimates for GMRES and consider the behaviour asymptotically as the mesh size h becomes small and the jump in coefficients becomes large. We observe, for our choice of Robin parameters, that increasing the jump in coefficients increases the convergence rate of GMRES. Numerical experiments are performed to verify the theoretical results.

AB - The optimised Schwarz method and the related two-Lagrange multiplier (2LM) method are nonoverlapping domain decomposition methods which can be used to numerically solve boundary value problems. Local Robin problems are solved on each subdomain in parallel to approximate the global solution, where a careful choice of Robin parameters leads to faster convergence. The 2LM method involves the solution of a nonsymmetric linear system that is usually solved with a Krylov subspace method such as GMRES. The speed of convergence of GMRES can be estimated using a conformal map from the exterior of the field of values of the system matrix to the interior of the unit disc. In this article we consider an elliptic PDE problem with a jump in diffusion coefficients across the interface between the subdomains. We approximate the field of values of the 2LM system matrix by a rectangle, R, in ℂ and provide optimised Robin parameters that ensure R is “well conditioned” in the sense that GMRES converges quickly. We derive convergence estimates for GMRES and consider the behaviour asymptotically as the mesh size h becomes small and the jump in coefficients becomes large. We observe, for our choice of Robin parameters, that increasing the jump in coefficients increases the convergence rate of GMRES. Numerical experiments are performed to verify the theoretical results.

KW - Domain decomposition

KW - GMRES convergence

KW - Heterogeneous media

KW - Optimised schwarz method

KW - Two-lagrange multiplier method

U2 - 10.1007/s11075-014-9923-z

DO - 10.1007/s11075-014-9923-z

M3 - Article

VL - 69

SP - 737

EP - 762

JO - Numerical Algorithms

JF - Numerical Algorithms

SN - 1017-1398

IS - 4

ER -