Titre :

Domain decomposition algorithms for two dimensional linear Schrödinger equation

Auteur(s) :

Besse, Christophe [Auteur]
Institut de Mathématiques de Toulouse UMR5219 [IMT]
Xing, Feng [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Maison de la Simulation [MDLS]

Titre de la revue :

Journal of Scientific Computing

Pagination :

735-760

Éditeur :

Springer Verlag

Date de publication :

2017

ISSN :

0885-7474

Mot(s)-clé(s) en anglais :

domain decomposition in space method AMS subject classifications 35Q55
Schrödinger equation
Schwarz waveform relaxation method
65M55
65Y05
65M60

Discipline(s) HAL :

Mathématiques [math]/Equations aux dérivées partielles [math.AP]
Mathématiques [math]/Analyse numérique [math.NA]

Résumé en anglais : [en]

This paper deals with two domain decomposition methods for two dimensional linear Schrödinger equation, the Schwarz waveform relaxation method and the domain decomposition in space method. After presenting the classical ...
Lire la suite >This paper deals with two domain decomposition methods for two dimensional linear Schrödinger equation, the Schwarz waveform relaxation method and the domain decomposition in space method. After presenting the classical algorithms, we propose a new algorithm for the free Schrödinger equation and a preconditioned algorithm for the general Schrödinger equation. These algorithms are studied numerically, which shows that the two new algorithms could accelerate the convergence and reduce the computation time. Besides the traditional Robin transmission condition, we also propose to use a newly constructed absorbing condition as the transmission condition.Lire moins >

Langue :

Anglais

Vulgarisation :

Non

Projet ANR :

Simulation numérique avancée pour les condensats de Bose-Einstein : Modèles numériques déterministes et stochastiques, Calcul haute performance, Simulation d'expériences physiques
Frontières, numérique, dispersion.

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL