Documents disponibles écrits par cet auteur

Large-scale parallel computation of incompressible flows by a domain decomposition-based least-squares finite element method / Xu Ding in Industrial & engineering chemistry research, Vol. 49 N° 17 (Septembre 1, 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 17 (Septembre 1, 2010) . - pp 8080–8085 Titre : Large-scale parallel computation of incompressible flows by a domain decomposition-based least-squares finite element method Type de document : texte imprimé Auteurs : Xu Ding, Auteur ; Q. Y. Jiang, Auteur ; Tate T. H. Tsang, Auteur Année de publication : 2010 Article en page(s) : pp 8080–8085 Note générale : Chimie industrielle Langues : Anglais (eng) Mots-clés : Incompressible  flows  Finite  element  method. Résumé : We have successfully developed a least-squares finite element method (LSFEM) for unsteady-state, three-dimensional convection dominated flows. The serial algorithm was used for simulations on a single CPU. In this work, our goal is to use a domain decomposition approach to parallelize the serial algorithm without much modifications of the serial code. We illustrate the efficiency of the parallel algorithm by flow simulations for more than 25 million unknowns. DEWEY : 660 ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie100136b [article] Large-scale parallel computation of incompressible flows by a domain decomposition-based least-squares finite element method [texte imprimé] / Xu Ding, Auteur ; Q. Y. Jiang, Auteur ; Tate T. H. Tsang, Auteur . - 2010 . - pp 8080–8085. Chimie industrielle Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 17 (Septembre 1, 2010) . - pp 8080–8085 Mots-clés : Incompressible  flows  Finite  element  method. Résumé : We have successfully developed a least-squares finite element method (LSFEM) for unsteady-state, three-dimensional convection dominated flows. The serial algorithm was used for simulations on a single CPU. In this work, our goal is to use a domain decomposition approach to parallelize the serial algorithm without much modifications of the serial code. We illustrate the efficiency of the parallel algorithm by flow simulations for more than 25 million unknowns. DEWEY : 660 ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie100136b