We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.NA

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Mathematics > Numerical Analysis

Title: Geometrical discretisations for unfitted finite elements on explicit boundary representations

Abstract: Unfitted (also known as embedded or immersed) finite element approximations of partial differential equations are very attractive because they have much lower geometrical requirements than standard body-fitted formulations. These schemes do not require body-fitted unstructured mesh generation. In turn, the numerical integration becomes more involved, because one has to compute integrals on portions of cells (only the interior part). In practice, these methods are restricted to level-set (implicit) geometrical representations, which drastically limit their application. Complex geometries in industrial and scientific problems are usually determined by (explicit) boundary representations. In this work, we propose an automatic computational framework for the discretisation of partial differential equations on domains defined by oriented boundary meshes. The geometrical kernel that connects functional and geometry representations generates a two-level integration mesh and a refinement of the boundary mesh that enables the straightforward numerical integration of all the terms in unfitted finite elements. The proposed framework has been applied with success on all analysis-suitable oriented boundary meshes (almost 5,000) in the Thingi10K database and combined with an unfitted finite element formulation to discretise partial differential equations on the corresponding domains.
Subjects: Numerical Analysis (math.NA); Computational Engineering, Finance, and Science (cs.CE)
DOI: 10.1016/j.jcp.2022.111162
Cite as: arXiv:2110.01378 [math.NA]
  (or arXiv:2110.01378v2 [math.NA] for this version)

Submission history

From: Santiago Badia Sb [view email]
[v1] Thu, 9 Sep 2021 04:43:04 GMT (6061kb,D)
[v2] Tue, 22 Mar 2022 23:42:22 GMT (7514kb,D)

Link back to: arXiv, form interface, contact.