This innovation addresses several critical issues in the construction of cut control volumes using embedded-boundary Cartesian mesh generators. The innovation considers the case of a surface triangulation intersecting the face of a Cartesian hex without intersecting any of its edges. Such face-piercing geometry is common in practical applications, especially when surface variations occur on a smaller scale than the Cartesian hex dimensions, i.e. coarse meshes. Examples include geometric perturbances such as nozzles, or even minor surface undulations and dimples. Furthermore, the innovation also considers the case of a surface triangulation containing internal passages within the Cartesian hex. The innovation correctly handles situations where hexes split into multiple control volumes are merged together if internal connections exist within the surface.
The innovation modifies several well known algorithms from the field of computational geometry and applies them to embedded-boundary Cartesian mesh generation. The resulting correct treatment of face-piercing geometry is critical for robust and accurate mesh generation and adaptation in the Cart3D software package (version 1.4 and subsequent releases).
NASA's Cart3D tool allows users to perform automated Computational Fluid Dynamics (CFD) analysis on a complex geometry. The package includes utilities for geometry import, surface modeling and intersection, mesh generation, flow simulation and post-processing of results. The main simulation code, flowCart, runs in parallel both in shared memory (OpenMP) and distributed memory (mpi) with excellent scalability. The package is highly automated so that geometry acquisition and mesh generation can usually be performed within a few minutes on most current desktop computers.
Geometry enters into Cart3D in the form of surface triangulations. These may be generated from within Computer-Aided Design (CAD) packages, from legacy surface triangulations, or from structured surface grids. Cart3D uses adaptively refined Cartesian grids to discretize the space surrounding geometry, and cuts the geometry out of the set of cut-cells that actually intersects the surface triangulation. The flow solver is parallel and can take full advantage of multi-core and multi-CPU hardware.