Automated Edge Grid Generation Based on Arc-Length Optimization

Abstract

Computational design and analysis has become a fundamental part of industry and academia for use in research, development, and manufac-turing. In general, the accuracy of a computational analysis depends heavily on the fidelity of the computational representation of a real-world object or phenomenon. Most mesh generation strategies focus on element quality–with the justification being that downstream applications require high quality ge-ometries in order to achieve a desired level of accuracy. However, element quality should be secondary to accurately representing the underlying physi-cal object or phenomenon. This work seeks to improve the process of creating a computational model of an object of interest by accelerating the process of mesh generation by reducing the need for (often) manual intervention. This acceleration will be accomplished by automatically generating optimal discretizations of curves by minimizing the arc-length deficit. We propose a method for generating optimal discretizations through local optimization of the arc length. Our results demonstrate the robustness and accuracy of our optimal discretization technique. We also discuss how to incorporate our edge grid generator into existing mesh generation software.