STTR Phase I: Parallel Lattice Kinetic Software for High Mach Number Fluid Dynamics

这个小型企业技术转移（STTR）第一阶段项目将生产一个先进的并行计算工具，用于统一预测连续流和稀薄流在现实世界中可重复使用的运载火箭的复杂性。该技术将是基于晶格玻尔兹曼方法（LBM）的数字物理技术的改进。该项目将首先在商用低马赫数/努森数流/声学模拟器PowerFLOW平台上开发和实施高马赫数模型。通过这种方式，将实现最高标准的数值精度/效率，并行可扩展性和几何灵活性。在对这种新的数值算法进行优化和基准测试后，研究团队将模拟一个全面的问题，包括在广泛的Knudsen范围内运行的空间飞行器的复杂几何形状，特别是马赫数。这里的目标是既要证明这种方法的并行计算效率，又要分析流动对飞行状态和设计变化的敏感性。如果这项概念验证工作取得成功，下一阶段的目标将是进一步开发这种并行计算工具，并在充分记录的全尺寸飞行器研究中对其进行测试，强调政府和私营企业感兴趣的飞行包线的空气动力学形状优化。该项目中统一的稀薄/连续流体预测工具有可能为扩展的PowerFLOW产品开辟新的主要商业市场，特别是在工程设计层面。首先，这项新技术将使航空航天工业中的高马赫数流预测成为可能。其次，由于底层数值模式的强局部性，这种新工具的并行效率将使使用成本效益高、基于cots的并行计算集群能够解决这些困难的流问题。第三，LBM方法解决可压缩流动问题的能力将为各种行业的新型LBM技术打开重要的新市场。最后，这项新技术将为计算机辅助工程开辟广阔的新市场，使车辆在非寻常的几何和流动物理条件下运行的形状优化成为可能，这些条件现在都是使用实验或半经验规则设计/优化的。

This Small Business Technology Transfer (STTR) Phase I project will produce an advanced parallel computational tool for unified prediction of continuum and rarefied flow over reusable launch vehicles of real world complexity. The technology will be an enhancement of the Digital Physics technology based on Lattice Boltzmann Methods (LBM). The project will start with development and implementation of a high Mach number model on the platform of commercial low Mach/Knudsen number flow/acoustics simulator PowerFLOW. In this way, the highest standards of numerical accuracy/efficiency, parallel scalability, and geometrical flexibility will be achieved. After optimizing and benchmarking this new numerical algorithm, the research team will simulate a full-scale problem including complex geometry of a space vehicle operating across a wide range of Knudsen and, especially, Mach numbers. The goal here will be to both demonstrate the parallel computational efficiency of this approach and to analyze the flow sensitivity to the flight regime and to design changes. If this proof-of-concept effort is successful, the goal of the next phase will be to further develop this parallel computational tool and test it on well-documented, full-scale vehicle studies emphasizing aerodynamic shape optimization for flight envelopes of interest to both Government and private industry.The unified rarefied/continuum flow prediction tool in this project has the potential to open major new commercial markets for the extended PowerFLOW product, especially at the engineering design level. First, this new technology shall enable high Mach number flow prediction within the aerospace industry. Secondly, this new tool's parallel efficiency, due to the strong locality of the underlying numerical schema, will enable the use of cost-effective, COTS-based, parallel computing clusters to solve these difficult flow problems. Thirdly, the ability of the LBM methods to address compressible flow problems should open important new markets for novel LBM-based technologies in a variety of industries. Finally, this new technology should open broad new markets for computer-aided engineering by enabling shape optimization of vehicles operating within nontrivial geometry and flow physics regimes, which are now designed/optimized using either experiment or semi-empirical rules.