Please use this identifier to cite or link to this item:
|Title:||Block Cartesian abstraction of a geometric model and its application in hexahedral mesh generation|
|Keywords:||Block Cartesian abstraction|
Hexahedral mesh generation
Laplacian- Isoparametric transformation
|Source:||Su, Y., Kumar, A.S. (2005-08). Block Cartesian abstraction of a geometric model and its application in hexahedral mesh generation. CAD Computer Aided Design 37 (9) : 899-907. ScholarBank@NUS Repository. https://doi.org/10.1016/j.cad.2004.09.017|
|Abstract:||In this work, a fuzzy logic approach is proposed to transform a geometric model of arbitrary shape to its block Cartesian abstraction. This abstraction is topologically similar to the original model and it contains geometric sub-entities which are all aligned in the Cartesian directions. This is achieved by calculating the modifications made to the face normal vectors as a result of the influences of the adjacent faces. A fuzzy logic inference engine is developed by combining heuristics to emulate the local changes in face normal vectors with respect to the changes in the global space. A three-dimensional field morphing algorithm is used to position the features of this block Cartesian abstraction so that a congruent geometric model can be reconstructed. Such a model is useful for the generation of structured quadrilateral boundary element meshes or structured hexahedral meshes based on grid-based meshing method, mesh mapping or sweeping. This approach is also able to overcome the traditional problem of having poorly shaped elements at the boundary using the grid-based method of mesh generation. As the topology of the block Cartesian abstraction is congruent to the original model, the mesh can be mapped back to the original model by employing an inverse operation of the transformation. © 2004 Elsevier Ltd. All rights reserved.|
|Source Title:||CAD Computer Aided Design|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Dec 11, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.