Anisotropic Cartesian Grid Generation Strategy for Arbitrarily Complex Geometry Based on a Fully Threaded Tree

Abstract

In this work, we present a powerful method for anisotropic Cartesian grid generation for arbitrarily complex 3D geometry. The proposed method addresses a number of challenging topics to achieve robustness and efficiency in mesh generation. In addition to displaying the basic information, including the generation of the initial uniform grid, the cell size, the cell number, and the cell centre coordinates, a modified fully threaded tree (FTT) data structure for use in the anisotropic Cartesian grid framework is also developed. Moreover, three criteria are proposed to ensure reasonable neighbourhood relations. A five-step adaptive mesh refinement strategy is developed to portray complex geometric details with high fidelity and maintain a reasonable number of cells. Three three-dimensional meshing experiments are implemented to illustrate and evaluate the present mesh generation performance, such as the number of cells, neighbour query runtime, total runtime, and memory consumption. Our results show that the proposed method overcomes the problems of a large number of cells and high memory consumption compared to the conventional isotropic Cartesian grid method. Additionally, mesh generation based on the modified FTT data structure outperforms the conventional 2 n tree data structure in terms of the neighbour query runtime, total runtime, and memory consumption. At last, a solution example is given to verify the proposed strategy. • A powerful strategy for anisotropic Cartesian grid generation is proposed. • A fully threaded tree (FTT) data structure for use in the anisotropic Cartesian grid framework is developed. • A five-step adaptive mesh refinement strategy is described. • The efficiency of anisotropic Cartesian grid generation for complex 3D geometry is assessed.