Loading…

A back-tracking procedure for optimization of simplex meshes

This work presents a back‐tracking procedure for the optimization of meshes. It can be applied to all ill‐shaped elements in a mesh in order to improve quality in mesh optimization. Basically, the original mesh is reconstructed in regions around the ill‐shaped elements by means of a set of deletion...

Full description

Saved in:
Bibliographic Details
Published in:Communications in numerical methods in engineering 2005-12, Vol.21 (12), p.711-722
Main Authors: Cavalcante-Neto, Joaquim B., Martha, Luiz Fernando, Wawrzynek, Paul A., Ingraffea, Anthony R.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This work presents a back‐tracking procedure for the optimization of meshes. It can be applied to all ill‐shaped elements in a mesh in order to improve quality in mesh optimization. Basically, the original mesh is reconstructed in regions around the ill‐shaped elements by means of a set of deletion and reconstruction operations based on visibility tests. The back‐tracking procedure fixes problems related to the quality of generated meshes in general and can be applied in two or three dimensions. The procedure works by deleting elements ranked below a predefined shape quality measure. It is important to mention that, although the presented procedure was devised for advancing‐front algorithms, it can also be used in meshes generated by different approaches, such as Delaunay and others, requiring no modification. The proposed procedure is applicable to simplex elements (triangles and tetrahedra), although ideally it could be extended to other types of elements. Examples of improved meshes using the back‐tracking procedure are presented, in which the quality of these meshes is assessed in order to validate the procedure proposed in this work. Copyright © 2005 John Wiley & Sons, Ltd.
ISSN:1069-8299
1099-0887
DOI:10.1002/cnm.786