Impact of triangle shapes using high-order discretizations and direct mesh adaptation for output error

Thesis (S.M.)--Massachusetts Institute of Technology, Computation for Design and Optimization Program, 2009. === This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. === Cataloged from student submitted PDF...

Full description

Bibliographic Details
Main Author: Sun, Huafei
Other Authors: David L. Darmofal.
Format: Others
Language:English
Published: Massachusetts Institute of Technology 2010
Subjects:
Online Access:http://hdl.handle.net/1721.1/54215
Description
Summary:Thesis (S.M.)--Massachusetts Institute of Technology, Computation for Design and Optimization Program, 2009. === This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. === Cataloged from student submitted PDF version of thesis. === Includes bibliographical references (p. 95-101). === The impact of triangle shapes, including angle sizes and aspect ratios, on accuracy and stiffness is investigated for simulations of highly anisotropic problems. The results indicate that for high-order discretizations, large angles do not have an adverse impact on solution accuracy. However, a correct aspect ratio is critical for accuracy for both linear and high-order discretizations. In addition, large angles are not problematic for the conditioning of the linear systems arising from discretization. They can be overcome through small increases in preconditioning costs. A direct adaptation scheme that controls the output error via mesh operations and mesh smoothing is also developed. The decision of mesh operations is solely based on output error distribution without any a priori assumption on error convergence rate. Anisotropy is introduced by evaluating the error changes due to potential edge split, and thus the anisotropies of both primal and dual solutions are taken into account. This scheme is demonstrated to produce grids with fewer degrees of freedom for a specified error level than the existing metric-based approach. === by Huafei Sun. === S.M.