Collaborative Research: Large Numerical Simulations of Turbulence and Reynolds, Schmidt and Rossby Number Scalings

建议编号：CTS-0553867/0553602PRINCIPAL研究员：P-K Yeung/K.R.SREENIVASANINTION：GIT/U，MD College PARKHIGH RESOLUSION数值模拟湍流本项目的目标是利用最高水平的超级计算能力，通过对大型数值模拟数据库的创新和严格分析，以及通过与合作者社区系统地共享数据，同时推动计算湍流的前沿，来促进对湍流的理解。重点讨论了局部涨落对时空尺度大小的依赖关系，以及几个主要的无量纲参数，它们表示了在固定和移动的参照系中各种尺度的影响、不同程度的分子扩散以及旋转坐标系中的科里奥利力的影响。PI将进行高分辨率(4096立方体或690亿个网格点的计算网格)模拟，并使用他们在模拟、理论和实验方面的综合专业知识来询问数据库，以提供关于湍流基本方面的明确答案，如间歇性、多重分形、各向异性、拉格朗日相似性、高施密特数下的混合、低Rossby数下的旋转湍流。此外，通过与一群备受尊敬的研究人员合作，数据库测试和激发的新想法的范围将大大扩大，他们将能够以无缝方式访问NSF TeraGrid的两个站点(SDSC和PSC)托管的数据。拟议的活动将产生广泛的影响，因为湍流在许多科学和工程领域的重要性，使用数百万超级计算机小时生成的模拟数据库的独特性，以及与至少两个由NSF支持的超级计算机中心的创新安排，以便与美国和海外不同兴趣的许多研究人员有效地共享数据。湍流科学的进步将对许多对社会重要的问题产生长期影响，包括化石燃料燃烧、海洋学中的海洋生物以及较新的技术，如纳米材料合成，所有这些都依赖于小尺度上的湍流混合。研究生将有独特和鼓舞人心的机会，通过与该领域许多杰出的研究人员密切合作，在科学计算的前沿工作，并发展更广阔的视角。类似的考虑预计将激励来自代表性不足群体的合格本科生。

PROPOSAL NO.: CTS-0553867 / 0553602PRINCIPAL INVESTIGATOR: P-K YEUNG/ K. R. SREENIVASANINSTITUTION: GIT / U. OF MD COLLEGE PARKHIGH RESOLUTION NUMERICAL SIMULATIONS OF TURBULENCE The objectives of this project are to utilize supercomputing power at the highest level to advance understanding of turbulent flows, via innovative and rigorous analyses of large numerical simulation databases, and via systematic sharing of data with a community of collaborators while pushing the frontiers of computational turbulence. Emphasis is placed on the dependence of local fluctuations on scale size in space and time, and on several major non-dimensional parameters that express the effects of a wide range of scales in both fixed and moving frames of reference, of different degrees of molecular diffusion, and of Coriolis forces in a rotating frame. The PIs will conduct high resolution (computational mesh of 4096 cubed, or 69 billion grid points) simulations and interrogate the database using their combined expertise in simulation, theory and experiment to provide definitive answers to fundamental aspects of turbulence, such as intermittency, multifractals, anisotropy, Lagrangian similarity, mixing at high Schmidt numbers, rotating turbulence at low Rossby numbers. Furthermore, the range of new ideas tested and stimulated by the database will be greatly expanded through working with a group of highly regarded researchers, who will be able to access the data hosted at two NSF TeraGrid sites (SDSC and PSC) in a seamless manner. The proposed activity will have wide impact because of the importance of turbulence in many fields of science and engineering, the uniqueness of the simulation database generated using millions of supercomputer hours, and innovative arrangements with at least two NSF-supported supercomputer centers for sharing data effectively with many researchers with varied interests in the US and abroad. Advances in the science of turbulent flow will have long-term impact in many problems important to society, including fossil fuel combustion, marine life in oceanography, and newer technologies such as nanomaterial synthesis, all of which depend on turbulent mixing at the small scales. Graduate students will have unique and inspiring opportunities to work at the forefront of scientific computing and develop broader perspectives by working closely with many distinguished researchers in the field. Similar considerations are expected to motivate qualified undergraduates from under-represented groups.