Virtual and smoothed finite elements: A connection and its application to polygonal/polyhedral finite element methods

Summary We show both theoretically and numerically a connection between the smoothed finite element method (SFEM) and the virtual element method and use this approach to derive stable, cheap and optimally convergent polyhedral FEM. We show that the stiffness matrix computed with one subcell SFEM is...

Full description

Saved in:
Bibliographic Details
Published inInternational journal for numerical methods in engineering Vol. 104; no. 13; pp. 1173 - 1199
Main Authors Natarajan, Sundararajan, Bordas, Stéphane PA, Ooi, Ean Tat
Format Journal Article
LanguageEnglish
Published Bognor Regis Blackwell Publishing Ltd 28.12.2015
Wiley Subscription Services, Inc
Subjects
Online AccessGet full text
ISSN0029-5981
1097-0207
DOI10.1002/nme.4965

Cover

More Information
Summary:Summary We show both theoretically and numerically a connection between the smoothed finite element method (SFEM) and the virtual element method and use this approach to derive stable, cheap and optimally convergent polyhedral FEM. We show that the stiffness matrix computed with one subcell SFEM is identical to the consistency term of the virtual element method, irrespective of the topology of the element, as long as the shape functions vary linearly on the boundary. Using this connection, we propose a new stable approach to strain smoothing for polygonal/polyhedral elements where, instead of using sub‐triangulations, we are able to use one single polygonal/polyhedral subcell for each element while maintaining stability. For a similar number of degrees of freedom, the proposed approach is more accurate than the conventional SFEM with triangular subcells. The time to compute the stiffness matrix scales with the O(dofs)1.1 in case of the conventional polygonal FEM, while it scales as O(dofs)0.7 in the proposed approach. The accuracy and the convergence properties of the SFEM are studied with a few benchmark problems in 2D and 3D linear elasticity. Copyright © 2015 John Wiley & Sons, Ltd.
Bibliography:istex:BC4FACF7E845470F5FE5783B205B8BF202FFD087
ArticleID:NME4965
ark:/67375/WNG-NKJMQ7QG-X
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0029-5981
1097-0207
DOI:10.1002/nme.4965