SBIR Phase I: An Automated Adaptive Cartesian/Prism Grid Flow Simulation Methodology for Arbitrary Moving Boundary Problems

点击翻译按钮获取中文摘要

*** 9660943 Wang This Small Business Innovation Research Phase I project will develop a new scientific-computation methodology for arbitrary moving-boundary problems. In the immediate vicinity of boundaries adaptive prism grids will be used in order to exploit their ability to provide surface-conformal spatial discretization, enabling accurate and efficient computational resolution of viscous and thermal boundary layers. These grids will move and deform with the boundaries around which they are built. Away from boundaries, a stationary, adaptive Octree-based Cartesian grid will be used to exploit its efficiency and flexibility. The Cartesian grid will be overlapped with the moving and deforming prism grids. Both types of grid will be adapted according to the physics of the unsteady flow field. For the coupling between the two grids, both conservative (expensive) and non-conservative (efficient) interfacing algorithms will be developed and demonstrated in the Phase I research. A second order flow solver based on a recently-developed all-speed flux splitting method capable of handing dynamic grids will be implemented, enabling highly-resolved solution of both incompressible and compressible flows. The unsteady flow fields will be displayed as they develop, through an on-line visualization capability. The overall system seamlessly integrates grid generation, flow solver, grid adaptation and post-processing to obtain maximum solution accuracy, efficiency and user friendliness. In Phase I, the methodology will be implemented and demonstrated in two dimensions. Extension to three-dimensions will be made in Phase II. The accurate modeling of flow phenomena involving moving boundaries is the key to a understanding, to improvement of performance, and to the cost-effective, rapid investigation of new designs. If successfully demonstrated in Phase I, the methodology will provide a reliable, computationally-efficient design and analysis tool that will have an immediate and far-reaching scient ific and economic impact in many distinct industries such as materials processing, aerospace engineering, and bioengineering. ***