Bulletin of the American Physical Society
67th Annual Meeting of the APS Division of Fluid Dynamics
Volume 59, Number 20
Sunday–Tuesday, November 23–25, 2014; San Francisco, California
Session L9: Minisymposium I: Frontiers of Computational Science in Transport Phenomena |
Hide Abstracts |
Chair: Gianluca Iaccarino, Stanford University Room: 3014/3016 |
Monday, November 24, 2014 3:35PM - 4:01PM |
L9.00001: Petascale Flow Simulations Using Particles and Grids Invited Speaker: Petros Koumoutsakos How to chose the discretization of flow models in order to harness the power of available computer architectures? Our group explores this question for particle (vortex methods, molecular and dissipative particle dynamics) and grid based (finite difference, finite volume) discretisations for flow simulations across scales. I will discuss methodologies to transition between these methods and their implementation in massively parallel computer architectures. I will present simulations ranging from flows of cells in microfluidic channels to cloud cavitation collapse at 14.5 PFLOP/s. [Preview Abstract] |
Monday, November 24, 2014 4:01PM - 4:27PM |
L9.00002: DNS/LES of Complex Turbulent Flows beyond Petascale Invited Speaker: Paul Fischer Petascale computing platforms currently feature million-way parallelism and it is anticipated that exascale computers with billion-way concurrency will be deployed by 2020. In this talk, we explore the potential of computing at these scales with a focus on turbulent fluid flow and heat transfer in a variety of applications including nuclear energy, combustion, oceanography, vascular flows, and astrophysics. Following Kreiss and Oliger '72, we argue that high-order methods are essential for scalable simulation of transport phenomena. We demonstrate that these methods can be realized at costs equivalent to those of low-order methods having the same number of gridpoints. We further show that, with care, efficient multilevel solvers having bounded iteration counts will scale to billion-way concurrency. Using data from leading-edge platforms over the past 25 years, we analyze the scalability of state-of-the-art solvers to predict parallel performance on exascale architectures. The analysis sheds light on the expected scope of exascale physics simulations and provides insight to design requirements for future algorithms, codes, and architectures. [Preview Abstract] |
Monday, November 24, 2014 4:27PM - 4:53PM |
L9.00003: High-Fidelity Simulations of Multiphysics Systems Invited Speaker: Frank Ham A pacing theme in the high-fidelity simulations of multi-physics flows is the continual push towards constitutive models that reflect the underlying physics more closely than ever before. At the same time, to impact the design and understanding of real fluidic devices, these models must ultimately be developed in the setting of a highly flexible computational infrastructure capable of both massive parallelism and geometric flexibility. This theme is illustrated using two multi-physics simulations that provide new incite into the behavior of complex fluidic devices. In the first, a novel unstructured Volume-of-Fluid (VoF) method is applied to simulate the liquid fuel atomization processes in a complex high shear nozzle typical of realistic gas turbine injectors. The simulation make aggressive use of directional grid adaptation to support the local resolution of critical instability mechanisms associated with the atomization process. In a companion example, the prediction of flow field and noise in a subsonic jet is linked critically to modeling and resolution of the nozzle boundary layers. [Preview Abstract] |
Monday, November 24, 2014 4:53PM - 5:19PM |
L9.00004: Computational aeroacoustics of turbulent high-speed jets Invited Speaker: Joseph W. Nichols Despite significant scientific investigation, jet noise remains a large component of the overall noise generated by supersonic aircraft. Experiments show that alterations to nozzle geometry, such as the addition of chevrons to the nozzle lip, can significantly reduce jet noise. In this talk, we assess unstructured large eddy simulation as a tool for predicting and understanding the aeroacoustic effects of complex geometry upon supersonic jets. Body-fitted, adaptive meshes are used to simulate the flow inside, around and through complicated nozzles, and results are validated against experimental measurements. High-fidelity simulations utilizing as many as one million processors simultaneously will be discussed, allowing for a detailed description of interactions between turbulence, shocks, and acoustics. This includes observations of the phenomenon of ``crackle'' noise in heated supersonic jets. We will briefly discuss challenges met and overcome along this frontier of com putational science, and describe how information extracted from the high-fidelity simulations can be used to construct accurate reduced-order models useful for aeroacoustic design. [Preview Abstract] |
Monday, November 24, 2014 5:19PM - 5:45PM |
L9.00005: Multiscale modeling of brain blow flow Invited Speaker: George Karniadakis Cardiovascular pathologies, such as brain aneurysms, are affected by the global blood circulation as well as by the local microrheology. Hence, developing computational models for such cases requires the coupling of disparate spatial and temporal scales often governed by diverse mathematical descriptions, e.g., by partial differential equations (continuum, 3D or 1D) and ordinary differential equations for discrete particles (atomistic). However, interfacing atomistic-based with continuum-based domain discretizations is a challenging problem that requires both mathematical and computational advances. We will present a physical model of the brain vasculature consisting at the macro level of all major arteries (about 200 down to 0.5 mm), at the mesoscale the fractal arteriolar tree (more than 10 millions down to 20 nm) and at the microscale the capillary bed. Correspondingly, we employ three different methods to model the total brain vasculature by developing proper interface conditions at each level. We will present examples from aneurysms and other hematological diseases, where red blood cell rheology is modeled explicitly. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700