The best of two worlds: The expedite boundary element method
► The method provides approximations of the matrices of the collocation BEM. ► Generally curved, 2D and 3D elements for potential and elasticity problems are used. ► No singular or quasi-singular numerical integration is necessary. ► The evaluation of results at any internal or boundary points requi...
Saved in:
Published in | Engineering structures Vol. 43; pp. 235 - 244 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
Kidlington
Elsevier Ltd
01.10.2012
Elsevier |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | ► The method provides approximations of the matrices of the collocation BEM. ► Generally curved, 2D and 3D elements for potential and elasticity problems are used. ► No singular or quasi-singular numerical integration is necessary. ► The evaluation of results at any internal or boundary points requires no integrations. ► The method can be advantageously used in combination with the fast multi-pole method.
The present developments result from the combination of the variationally-based, hybrid boundary element method and a consistent formulation of the conventional, collocation boundary element method. The procedure is simple to implement and turns out to be computationally faster than the mentioned, preceding numerical methods – and almost as accurate – for the analysis of large-scale, two-dimensional and three-dimensional problems of potential and elasticity of general shape and topology, also applicable to time-dependent problems. Both the double-layer and the single-layer potential matrices of the collocation boundary element method, H and G, respectively, whose standard evaluation requires dealing with singular and improper integrals, are obtained in an expedite way that circumvents almost any numerical integration – except for a few regular integrals. Since the resultant matrices do not differ in nature from the ones of the conventional, collocation boundary element method, the developments are suited for a matrix solution in terms of a GMRES algorithm, for example, and in the framework of the fast multi-pole method, so that very large problems can be ultimately dealt with efficiently. A few numerical examples are shown to assess the applicability of the method, its computational effort and some convergence issues. |
---|---|
Bibliography: | ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23 ObjectType-Article-1 ObjectType-Feature-2 |
ISSN: | 0141-0296 1873-7323 |
DOI: | 10.1016/j.engstruct.2012.04.042 |