Return to search

Optimized Schwarz methods for the advection-diffusion equation

The optimized Schwarz methods were recently introduced to enhance the convergence of the classical Schwarz iteration, by replacing the Dirichlet transmission conditions with different conditions obtained through an optimization of the convergence rate. This is formulated as a min-max problem. These new methods are well-studied for elliptic second order symmetric equations. The purpose of this work is to compute optimized Robin transmission conditions for the advection-diffusion equation in two dimensions, by finding the solution of the min-max problem. The asymptotic expansion, for small mesh size h, of the resulting convergence rate is found: it shows a weak dependence on h, if the overlap is 0(h) or no overlap is used. Numerical experiments illustrate the improved convergence of these optimized methods compared to other Schwarz methods, and also justify the continuous Fourier analysis performed on a simple model problem only. The theoretical asymptotic performance is also verified numerically.

Identiferoai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.19701
Date January 2003
CreatorsDubois, Olivier
PublisherMcGill University
Source SetsLibrary and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada
LanguageEnglish
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Formatapplication/pdf
CoverageMaster of Science (Department of Mathematics and Statistics)
RightsAll items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated.
Relationalephsysno: 002022786, Theses scanned by McGill Library.

Page generated in 0.0015 seconds