Clustering: Automated detail suppression using virtual topology

Alla Sheffer*, Ted Blacker, Michel Bercovier

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

29 Scopus citations

Abstract

Model adjustment for meshing is one of the most cumbersome steps in generation of a finite-element mesh. While many different methods for automatic mesh generation exist, the adjustment of the model so it will be suitable for those methods, remains time consuming and often requires extensive user interaction. In this work a new automated method for model simplification is suggested. The method uses virtual topology concept as introduced in our previous work, and provides the means to simplify the model topology so it will be suitable for meshing, this without changing the geometry. The algorithm is based on clustering of planar facets approximating the model faces into regions, based on angle and distance criteria. Then based on resulting clusters analysis, the model faces are virtually merged or collapsed, achieving topological simplification of the model. The method is based on computations of plane relations and therefore is very efficient.

Original languageEnglish
Pages (from-to)57-64
Number of pages8
JournalAmerican Society of Mechanical Engineers, Applied Mechanics Division, AMD
Volume220
StatePublished - 1997

Fingerprint

Dive into the research topics of 'Clustering: Automated detail suppression using virtual topology'. Together they form a unique fingerprint.

Cite this