Improved Mesh Morphing Based on Radial Basis Functions

Abstract

CFD is a major aspect of the work by the German Aerospace Center. In order to develop an optimized flow configuration it is necessary to minimize drag. Thereto different shape variations and meshes have to be investigated. In the CFD process two kinds of realizations exist. One possibility is to create a new mesh for every configuration. The other is mesh morphing. A lot of additional time is needed to generate new CAD description and generating a mesh, even more doing the calculations from scratch is time consuming. Mesh morphing is a better solution to investigate shape variations, it means shifting the position of points in the grid. So the adapted mesh has the same cell topology as the old one. Since the old solution can be used for new computations the work flow can be accelerated. Therefore, a mesh morphing tool has been developed to create variations of existing meshes. This tool uses radial basis functions (RBF) because they are fast and stable. RBF are used to interpolate morphing vectors on scattered data based on given control points. By moving cell points in cell structure it might happen that cells become distorted. A new methodology to handle this has been invented.