Bulletin of the American Physical Society
65th Annual Meeting of the APS Division of Fluid Dynamics
Volume 57, Number 17
Sunday–Tuesday, November 18–20, 2012; San Diego, California
Session R27: Computational Fluid Dynamics IX |
Hide Abstracts |
Chair: Onkar Sahni, Rensselaer Polytechnic Institute Room: 31C |
Tuesday, November 20, 2012 1:00PM - 1:13PM |
R27.00001: Non-isothermal 3D SDPD Simulations Jun Yang, Raffaele Potami, Nikolaos Gatsonis The study of fluids at micro and nanoscale requires new modeling and computational approaches. Smooth Particle Dissipative Dynamics (SDPD) is a mesh-free method that provides a bridge between the continuum equations of hydrodynamics embedded in the Smooth Particle Hydrodynamics approach and the molecular nature embedded in the DPD approach. SDPD is thermodynamically consistent, does not rely on arbitrary coefficients for its thermostat, involves realistic transport coefficients, and includes fluctuation terms. SDPD is implemented in our work for arbitrary 3D geometries with a methodology to model solid wall boundary conditions. We present simulations for isothermal flows for verification of our approach. The entropy equation is implemented with a velocity-entropy Verlet integration algorithm Flows with heat transfer are simulated for verification of the SDPD. We present also the self-diffusion coefficient derived from SDPD simulations for gases and liquids. Results show the scale dependence of self-diffusion coefficient on SDPD particle size [Preview Abstract] |
Tuesday, November 20, 2012 1:13PM - 1:26PM |
R27.00002: Comparison study of meshfree methods for viscous flow Zhenyu He, Louis Rossi We compare and contrast two meshfree schemes for viscous flow: Smoothed particle hydradynamics (SPH) and vortex methods (VM). SPH and VM are widely used meshfree particle in fluid dynamic applications. SPH is more flexible for capturing multiphysics problems. VM is better developed theoretically but has a more limited scope of applications. In SPH, the state of fluid system is represented by a set of moving basis functions which represent material properties such as density and momentum. Vortex particle methods represent a discretization of the vorticity field and use a Greens kernel to determine the velocity field. Our aim is to clarify the role played by the most commonly used viscous terms in SPH and VM in simulating incompressible fluid flow. Special test problems are used in order to remove the boundary effect to the results. We will present the accuracy and the efficiency of the different schemes which highlight the importance of key parameters in the algorithms including core width, overlap and equations of state. [Preview Abstract] |
Tuesday, November 20, 2012 1:26PM - 1:39PM |
R27.00003: Scale-bridging schemes based on the material point method Shaolin Mao, Xia Ma, Virginie DuPont, Duan Zhang With recent development of heterogeneous computational resources, such as combined GPU and CPU computations, there is an emerging possibility to apply a continuum approach to thermodynamically non-equilibrium systems with the closure quantities, such pressure, computed directly from numerical simulation of systems at a smaller length and time scales. Although it may not be possible to calculate the entire physical system at the small length and time scales, one can calculate closure quantities at representative locations, such as Gauss points in a finite element calculation, or mesh nodes, cell centers in a finite volume calculation, by surrounding those points with small volumes, called sub-systems, and perform directly numerical simulation on them to consider physical interactions at the smaller time and length scales. Before this hopeful method can be practical, we need to study issues related history dependency, time interval and spatial size of the sub-systems to simulate in order to calculate the closure quantities with credible accuracy. We also need to study methods to communicate between the sub-systems. For this purpose, we develop a numerical scheme base on the material point method (MPM). Results from such studies and from the numerical scheme will be presented. [Preview Abstract] |
Tuesday, November 20, 2012 1:39PM - 1:52PM |
R27.00004: Artificial Compressibility with Entropic Damping Jonathan Clausen, Scott Roberts Artificial Compressibility (AC) methods relax the strict incompressibility constraint associated with the incompressible Navier--Stokes equations. Instead, they rely on an artificial equation of state relating pressure and density fluctuations through a numerical Mach number. Such methods are not new: the first AC methods date back to Chorin (1967). More recent applications can be found in the lattice-Boltzmann method, which is a kinetic/mesoscopic method that converges to an AC form of the Navier--Stokes equations. With computing hardware trending towards massively parallel architectures in order to achieve high computational throughput, AC style methods have become attractive due to their local information propagation and concomitant parallelizable algorithms. In this work, we examine a damped form of AC in the context of finite-difference and finite-element methods, with a focus on achieving time-accurate simulations. Also, we comment on the scalability of the various algorithms. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. [Preview Abstract] |
Tuesday, November 20, 2012 1:52PM - 2:05PM |
R27.00005: Solving Navier-Stokes' equation using Castillo-Grone's mimetic difference operators on GPUs Mohammad Abouali, Jose Castillo This paper discusses the performance and the accuracy of Castillo-Grone's (CG) mimetic difference operator in solving the Navier-Stokes' equation in order to simulate oceanic and atmospheric flows. The implementation is further adapted to harness the power of the many computing cores available on the Graphics Processing Units (GPUs) and the speedup is discussed. [Preview Abstract] |
Tuesday, November 20, 2012 2:05PM - 2:18PM |
R27.00006: Mathematical modelling of backflushing in membrane separation Frank Vinther A mathematical model of pressure driven membrane separation is presented. Concentration polarization is well known to reduce the flux, and thereby the performance of the separation process, due to an increase in osmotic pressure. Therefore, back shocking where the pressure difference and the flow through the membrane is reversed, is known to decrease the concentration polarization and increase the performance of the membrane. The interplay between back shocking amplitude, frequency and reduction in concentration polarization at the membrane surface is investigated. [Preview Abstract] |
Tuesday, November 20, 2012 2:18PM - 2:31PM |
R27.00007: Size-Dependent Couple-Stress Fluid Mechanics and Application to the Lid-Driven Square Cavity Flow Arezoo Hajesfandiari, Gary Dargush, Ali Hadjesfandiari We consider a size-dependent fluid that possesses a characteristic material length $l$, which becomes increasingly important as the characteristic geometric dimension of the problem decreases. The term involving $l$ in the modified Navier-Stokes equations \[ \rho \frac{D{\rm {\bf v}}}{Dt}=-\nabla p+\mu \nabla ^2{\rm {\bf v}}-\mu l^2\nabla ^2\nabla ^2{\rm {\bf v}} \] generates a new mechanism for energy dissipation in the flow, which has stabilizing effects at high Reynolds numbers. Interestingly, the idea of adding a fourth order term has been introduced long ago in the form of an artificial dissipation term to stabilize numerical results in CFD methods. However, this additional dissipation has no physical basis for inclusion in the differential equations of motion and is never considered at the boundary nodes of the domain. On the other hand, our couple stress-related dissipation is physically motivated, resulting from the consistent application of energy principles, kinematics and boundary conditions. We should note, in particular, that the boundary conditions in the size-dependent theory must be modified from the classical case to include specification of either rotations or moment-tractions. In order to validate the approach, we focus on the lid-driven cavity problem. [Preview Abstract] |
Tuesday, November 20, 2012 2:31PM - 2:44PM |
R27.00008: Fluctuating hydrodynamics for coarse-grained implicit solvent models of soft materials Paul Atzberger Many coarse-grained models have been developed and effectively employed in equilibrium studies of soft materials by treating implicitly interactions mediated by the solvent. These include models for polymeric materials, gels, and lipid bilayer membranes. However, dynamic studies require at a minimum incorporating momentum transfer through the solvent degrees of freedom. Hydrodynamic theories often give a sufficient kinetic description of the solvent mediated coupling of microstructures. In soft materials such theories must be reconciled with thermal fluctuations and entropic effects that play a significant role. In this talk we discuss new approaches based on fluctuating hydrodynamics to incorporate such features into implicit solvent coarse-grained models. For efficient simulations of the solvent-microstructure interactions and thermal fluctuations, we introduce new numerical methods based on the Stochastic Eulerian Lagrangian Method (SELM). We then discuss results for specific coarse-grained models of polymeric fluids, gels, and lipid bilayer membranes. [Preview Abstract] |
Tuesday, November 20, 2012 2:44PM - 2:57PM |
R27.00009: Development of a Computational Tool for Inductively Coupled Plasma Flow Over Test Samples Maximilian Dougherty, Douglas Fletcher A boundary layer flow code is being developed to complement experimental work at the University of Vermont Plasma Diagnostics Laboratory. The stagnation region boundary layer is important because it controls the heat flux to the material during planetary entry and ground testing. Within the nonequilibrium boundary layer, highly exothermic chemical reactions can significantly augment heat transfer to the material surface. Many boundary layer codes rely on similarity solutions in transformed coordinate systems that are not necessarily intuitive at first glance. The benefit of these transformations is that a simplified grid can be used with a finite difference approach but this is at the expense of obfuscating the basic physical quantities which are of interest from an experimental perspective. As a result, the code being developed will use a finite volume formulation in physical coordinates. Chemical and thermodynamic properties will be computed with a separate gas property library which will allow for calculations of high temperature dissociated multi-species flows. [Preview Abstract] |
Tuesday, November 20, 2012 2:57PM - 3:10PM |
R27.00010: Coalescence of two colliding liquid droplets with lattice Boltzmann method Yikun Wei, Yuehong Qian In this talk, we present two-dimensional numerical simulations of the head-on and off-center binary collision of van der Waals liquid droplets using the lattice Boltzmann method (LBM). The effects of the Weber number, impact velocity, droplet size ratio on the coalescence process are investigated. Numerical results are found in a good agreement with experimental findings.\\[4pt] [1] Y.H. Qian, D. d'Humieres, and P. Lallemand, Europhys. Lett. 17, 479 (1992).\\[0pt] [2] X.W. Shan and H.D. Chen, Lattice Boltzmann model for simulating flows with multiple phases and components, Physical Review E 47 (3) (1993) 1815--1819.\\[0pt] [3] M.R. Swift, W.R. Osborn, and J.M. Yeomans, Lattice Boltzmann simulation of nonideal fluids, Physical Review Letters 75 (5) (1995) 830--833.\\[0pt] [4] Y.K. Wei Y.H. Qian, Lattice Boltzmann Simulations for Multiphase Fluids with the Redich-Kwong Equation of State, Journal of Hydrodynamics, 2011,23(6)823-828 [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