Engineers using computational fluid dynamics (CFD) to simulate gas and fluid flows in their designs often experience a number of limitations using current tools. The primary limitation is computational capacity, as even high performance computing (HPC) systems are often unable to perform meshing and CFD to a desired level of precision within a reasonable timeframe. NASA refers to this lack of scalable CFD pre- and post-processing methods in “CFD Vision 2030 Study: A Path to Revolutionary Computational Aerosciences.” We have learned to live with decreasing meshing precision, simulating simplified subsystems, and waiting hours or days for results as we struggle with this limitation. While we all wish for the practical “virtual wind tunnel,” it remains an elusive wish for a future time.
Engineers can find encouragement in ongoing efforts to improve meshing algorithms that can help address this issue. HPC innovations will contribute to the solution, but new algorithms and approaches are necessary to achieve significant performance improvements. Pointwise, Inc., is focusing on the high-fidelity mesh generation aspect of this problem and has been beta testing the Project Geode mesher for the last year with organizations such as Stanford University’s SU2 team. Pointwise is now seeking engineers who can participate in beta testing to further improve and validate the tool.
Engineers can contribute toward solving this problem by helping to beta test the latest revision of the Pointwise Project Geode mesher, a fourth-generation meshing kernel optimized for engineering simulation. Beta testers will help to validate that Project Geode provides the needed capability required by CFD solvers for tasks like mesh adaptation, as well as delivers the capacity needed to complete meshing in usable time frames. Engineers can also help influence tool features and capabilities by providing user feedback to improve the product. Engineers can participate in the beta testing at no cost, but will need to execute a beta agreement.
Engineers can use the kernel on most computers, and Windows, Linux and Mac operating systems are supported for both 32- and 64-bit systems. Users can generate structured, unstructured and hybrid meshes, and export data in formats compatible with most CFD solvers, including ANSYS, STAR-CCM+, OpenFOAM and SU2, as well as vendor-neutral formats such as CGNS. Automated mesh generation is supported with the Glyph scripting language.