Hexahedral Mesh Generation Algorithm Based On Feature Constraints

Finite element mesh generation is the first step of the numerical analysis methods, as well as the the necessary premise of application mechanism and process model simulation and analysis to the geological calculation. The quality of mesh will directly affect the numerical analysis and study the accuracy and efficiency of calculation. So the finite element mesh generation method is a long-time hot topic. Among all the mesh element types, the hexahedron has been widely used because of having a distinct advantage in the quantity, precision and anti distortion relativing to the tetrahedron. However, the research of hexahedral mesh generation method is not as good as tetrahedral mature. The method of hexahedral mesh generation still exists many technical problems to be solved.Traditional hexahedral mesh generation method considered less in the complex feature constraints, insufficiency in the design for the algorithm of feature constraint. In fact, feature constraints are an important class of objects in the actual existence of the object of study in geoscience, the support on complex feature constraints is also a essential factors of a good hexahedral mesh generation. This paper focus on the core problem of high quality hexahedral mesh generation with constraints based on complex features. By generating feature classification and mapping to feature constraints data set as the preceding data of hexahedral mesh generation. Based on the preceding data, this paper studies the method of generating structured and unstructured mesh, gives the feature constraint processing template as the basis of inquiry optimization method of element quality to generate the hexahedral mesh on considering of the complex feature constraints. The main results and conclusions of this paper are as follows:(1) The configuration and mapping mechanism of feature constraints. This paper gives the definition of hexahedral mesh feature constraint based on the traditional concept of constraint and the feature of learning object. The classification of the feature constraint is according to the dimension, acquisition methods and the effects on the mesh generation. By mapping the feature constraints generation and match the mesh preceding data set, using feature constraints as a mesh of the data source, let the treatment of feature constraints and mesh generation closely connected.(2) Hexahedral mesh generation methods that considering the complicated feature constraints. By setting different geological objects scenes, comparing to analysis different variety of hexahedral mesh generating methods, this paper summarizes the improvement on the basis of the existing mesh generation method. Put forward two kinds of mesh generation method:modified sweeping method and the receding front method, meshing the study area of complex feature constraints. Considering the surface and internal complex feature constraints, this paper generate considering feature constraints structured hexahedral mesh and unstructured hexahedral mesh.(3) The processing methods for feature constraints in the hexahedral mesh. The post-processing of feature constraints in the hexahedral mesh (point, line, surface, volume) is according to its feature of huge data amount and the uneven spatial distribution. Mainly through the method of template and surface fitting for processing complex feature constraint which the preceding cannot handle. To achieve precise fitting feature constraints, generation the finite element mesh to meet feature constraints’ distribution. Optimizes the quality of mesh element aiming at the mesh element nodes displacement constraints adjustment and the feature of the treatment process while using topological relationship optimization and geometry optimization method.