A Tri-Tree multigrid recoarsement algorithm for the finite element formulation of the Navier-Stokes equations

In the present paper local recoarsements of finite element grids for solving the Navier-Stokes equations are investigated. The recoarsement algorithm is based on the hierarchic Tri-Tree multigrid generation algorithm and computes a local Tri-Tree Element Reynolds number. Previous investigations show...

Full description

Saved in:
Bibliographic Details
Published in:Computer methods in applied mechanics and engineering Vol. 135; no. 1; pp. 129 - 142
Main Author: Wille, S.Ø.
Format: Journal Article
Language:English
Published: Amsterdam Elsevier B.V 15.08.1996
Elsevier
Subjects:
Computational methods in fluid dynamics
Exact sciences and technology
Fluid dynamics
Fundamental areas of phenomenology (including applications)
Physics
Finite element method
Numerical simulation
Algorithms
Multigrid
Navier-Stokes equations
ISSN:0045-7825, 1879-2138
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the present paper local recoarsements of finite element grids for solving the Navier-Stokes equations are investigated. The recoarsement algorithm is based on the hierarchic Tri-Tree multigrid generation algorithm and computes a local Tri-Tree Element Reynolds number. Previous investigations show that the Element Reynolds number is critical for convergence of the linear CGSTAB equation solver. A necessary condition for convergence is found to be that the Element Reynolds number is less than 10. The Tri-Tree multigrid solution algorithm starts by solving the equations for a low Reynolds number for a coarse grid. Solutions for higher Reynolds numbers are obtained by gradually increasing the velocity boundary conditions while refining the grid correspondingly to satisfy the Element Reynolds criterion. As the global Reynolds number is increasing, the non-linear convection moves in space. The recoarsement algorithm is then making the grid coarser at locations where the previous fine grid resolution is no longer required.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0045-7825
1879-2138
DOI:10.1016/0045-7825(95)00966-3