Open Access

ARTICLE

A New Anisotropic Local Meshing Method and Its Application in Parametric Surface Triangulation

by W.W. Zhang¹, Y.F. Nie¹, Y.Q. Li¹

1 Department of Applied Mathematics, School of science, Northwestern Polytechnical University, Xi’an 710129, PR China. E-mail: yfnie@nwpu.edu.cn

Computer Modeling in Engineering & Sciences 2012, 88(6), 507-530. https://doi.org/10.3970/cmes.2012.088.507

Abstract

A new algorithm for anisotropic triangular meshes generation in two dimension is presented. The inputs to the algorithm are the boundary geometry information and a metric tensor that specifies the desired element size and shape. The initial nodes are firstly distributed according to the above mentioned geometrical information, after bubble simulation, the optimized nodes set that meets the requirements of the metric tensor is obtained quickly. Then taking full advantage of the nodes set and the adjacency lists information provided by the process of node placement, a handful of non-satellite nodes are removed from the adjacency lists of the nodes with the Anisotropic Bubble-type Local Mesh Generation (ABLMG) method, the anisotropic meshes are generated rapidly. The developed meshes reflect the metric tensor requirement well, and they don't require remeshing and smoothing. Since the adjacency list is built for each node, it avoids searching neighbor nodes when calculating the resultant force of each node and generating the local mesh around each node, the simulation time is greatly saved. The new algorithm is also used for 3D surface triangulation based on mapping method. The parametric space corresponding to the target surface is triangulated with our algorithm, then mapping the mesh topology in the parametric space back to the original surface, the surface mesh with high quality are obtained. Some numerical examples are given to test the feasibility of the algorithm.

---

Keywords

---