In this talk we describe what we mean by the seeming oxymoron, Cartesian grid generation. By this terminology, we mean the representation of solid objects using embedded surface descriptions superimposed on a regular Cartesian grid. This method has the advantage that no explicit mesh generation is needed, greatly reducing the human effort involved in complex flow computations. Our algorithms allow for component-based geometry specification, so that components can be translated with respect to each other without user intervention. We pre-process the geometry to account for these intersections before proceeding to the volume mesh generation.

Back to Workshop Schedule

1996-1997 Mathematics in High Performance Computing