Research On Structure Simplification Approach Of Hexahedral Mesh In CAE

In the field of hexahedral mesh generation,generation of valid and coarse hexahedral meshes with a simple singularity structure is a challenging problem.This type of mesh plays an extremely important role in the field of meshing and isogeometric analysis.Based on these requirements,a large number of hexahedral mesh structure simplification approaches are constantly proposed,which achieve the purpose of indirectly generating the meshes with simplified structure by performing iterative simplification processes.Although these researches have been applied well in the field of graphics,there are still some obvious defects,which includes the simplification targets are ambiguous in the pipeline,the convergence speed is slow and the error of boundary shape is hard to control.How to overcome these defects becomes a meaningful work.In the paper,the existing hexahedral mesh simplification methods in CAE are discussed,and a novel singularity structure simplification approach based on weighted ranking method is proposed.Singularity structure simplification of hexahedral mesh by weighted ranking approach is based on the base-complex structure.The iterative collapsing operation based on base-complex structure and base-complex component is used to eliminate singularity point step by step,and the element quality and shape error are improved by local parameterization.The simplified mesh is generated by above operations.The main contributions can be summarized as follows:1.By analyzing the distribution of singularities,we proposed a weighted ranking algorithm for the pipeline of simplification process,which can improve the convergence speed of simplification effectively,and rapidly reduce the number of sheets with close-loop or self-intersection configuration.In addition,our method can greatly improve the reduction rate of components while preventing early termination of the pipeline.2.The local parameterization SLIM method is used in this method,and an optimization method based on SIM framework is proposed,which can significantly improve the uniformity of elements,and can accurately maintain the uniformity of the region around the feature lines and sharp features.3.Based on the calculation of the shape error,a series of local refinement rules is proposed,and these rules are used to control the number of elements to the specified target precisely while maintaining the uniformity and reducing the shape error.In addition,these refinement rules can delay the occurrence of geometric constraints,and improve the reduction rate of components.A common data set is experimented.The data set mainly includes the meshes generated by Octree-based method,Polycube-based method and frame-field method.we compared our approach with several mature methods in recent years,the results show that our approach can effectively improve the mesh quality and reduction ratio.