Bulletin of the American Physical Society
2005 58th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 20–22, 2005; Chicago, IL
Session AG: CFD I |
Hide Abstracts |
Chair: Tim Colonius, California Institute of Technology Room: Hilton Chicago Williford A |
Sunday, November 20, 2005 8:00AM - 8:13AM |
AG.00001: Insights into efficient laminar mixing from studies on the gastrointestinal tract Seth Dillard, Sreedevi Krishnan, H.S. Udaykumar Many biological and engineered systems are dependent upon the efficacy of fluid mixing in the laminar flow regime. The physiological processes responsible for nutrient absorption and transport in the GI tract are prime examples of natural design optimization from which much can be learned and applied to designs of engineering interest. We have analyzed individual aspects of the GI tract, with particular focus on the junction between the stomach and intestine to quantify the effect of the anatomy and physiology of this segment on the mixing of nutrients with enzymes. An Eulerian levelset-based flow solver with Lagrangian particle tracking is used to computationally analyze fluid flow through several representative 2-D components of the antral-duodenal junction. Stretch rate maps and scalar species distributions illustrate how each individual anatomical feature and peristaltic contraction patterns contribute to mixing through vortex formation and shedding in the presence of pulsatile flow. It is shown that geometric irregularities and asymmetries, such as those found in biological systems, can lead to enhanced fluid mixing. [Preview Abstract] |
Sunday, November 20, 2005 8:13AM - 8:26AM |
AG.00002: Simulation of the motion of elliptic bodies in a viscous incompressible fluid L. Hector Juarez In this work we discuss the application of a methodology combining distributed Lagrange multiplier based fictitious domain techniques, finite element approximation, and operator splitting, to the numerical simulation of the motion of elliptic bodies in an incompressible viscous fluid. The interaction between the fluid and the rigid body is implicitly modeled by a global variational formulation, so that we do not compute the hydrodynamical forces explicitly during the simulation. In addition, the fictitious domain method allows the flow computation to be done on a fixed region without re-meshing. Examples of the sedimentation of a elliptic body, and of hydrodynamic pendula will be presented. [Preview Abstract] |
Sunday, November 20, 2005 8:26AM - 8:39AM |
AG.00003: Fourth Order Compact Formulation of Navier-Stokes Equations and Driven Cavity Flow at High Reynolds Numbers Ercan Erturk, Cihan Gokcol, Bahtiyar Dursun, Hakan Kaykisizli A new fourth order compact formulation for the steady 2-D incompressible Navier-Stokes equations is presented. The uniqueness of this formulation is that the final form of the fourth order compact formulation is in the same form of the Navier-Stokes equations such that any numerical method that solve the Navier-Stokes equations can be easily applied to this fourth order compact formulation. Moreover, with this formulation, any existing code that solve the Navier-Stokes equations with second order accuracy ($\mathcal O$$\Delta x^{2}$) can easily be altered to provide fourth order accurate ($\mathcal O$$\Delta x^{4}$) solutions just by adding some coefficients into the code at the expense of extra CPU work of evaluating these coefficients. The efficiency of this formulation will be demonstrated. [Preview Abstract] |
Sunday, November 20, 2005 8:39AM - 8:52AM |
AG.00004: Accuracy assessment of inverse-distance weighted interpolation in the immersed boundary method Yu-heng Tseng, Gao Tong Inverse-distance weighted (IDW) interpolation has been recommended in the immersed boundary method. The interpolation is simple and flexible. However, its accuracy is still not clear yet. We establish a theorem for the error estimation of the IDW interpolation and assess its numerical accuracy on a Cartesian mesh. The accuracy is evaluated and discussed using several analytical functions. Some numerical examples are presented to illustrate the performance and validate the accuracy using a ghost-cell immersed boundary method. These includes flow past a circular cylinder, flow over a sphere, and oceanic flow around an island. These results show significant improvement in terms of the accuracy and stability. [Preview Abstract] |
Sunday, November 20, 2005 8:52AM - 9:05AM |
AG.00005: Application of FCT to Incompressible Flows Junhui Liu, David Mott, Carolyn Kaplan, Elaine Oran Flux-corrected transport, FCT, has been applied to incompressible flows on a collocated grid. LCPFCT, a standard version of FCT, has been used to calculate the advection term in the momentum equation. The flux limiter embedded in the FCT algorithm helps to stabilize oscillations introduced by the discretization of the advection term. It is found that if the pressure-gradient term is included in the computation of intermediate velocities, the discretization of the pressure gradient term introduces odd-even decoupling to the discretization of the Poisson equation, and results in pressure oscillations. Therefore, to avoid this odd-even decoupling problem, the pressure gradient term is either not included in the calculation of the intermediate velocities, or is removed from the intermediate velocities after the integration of the momentum equations. This extension of the FCT algorithm to incompressible flow provides us with a time accurate and robust approach. It gives satisfactory results to a variety of problems over a wide range of velocity and Reynolds number. [Preview Abstract] |
Sunday, November 20, 2005 9:05AM - 9:18AM |
AG.00006: Immersed Boundary Fractional Step Method Kunihiko Taira, Tim Colonius We present a new formulation of the immersed boundary method for incompressible flow over moving rigid bodies. Like many existing techniques we introduce a set of interpolation points on the surface at which the no-slip boundary condition is satisfied by including a (regularized) force in the momentum equations. By introducing interpolation and regularization operators and grouping pressure and force unknowns together, the discretized Navier-Stokes equations with the immersed boundary method can be formulated with an identical structure to the traditional fractional step method, but with a modified Poisson equation whose unknowns are both the pressure and the boundary force. The method highlights the analogous roles of pressure and boundary forcing as Lagrange multipliers in order to satisfy the divergence free and no-slip constraints, respectively. The overall method is found to be a simple addition to an existing fractional step code and the extended Poisson equation is solved efficiently with the conjugate gradient method. We demonstrate convergence and present results for two-dimensional flows with a variety of moving rigid bodies. [Preview Abstract] |
Sunday, November 20, 2005 9:18AM - 9:31AM |
AG.00007: A 3D Immersed Interface Method Sheng Xu, Jane Wang We present the 3D implementation of an immersed interface method (IIM) which is capable of simulating multiple moving rigid or flexible objects in an unsteady fluid. The IIM is a variant of the immersed boundary method, which solves the Naviers-Stokes equations subject to singular force. The difference between the two methods is that the IIM handles the singular force in terms of the jump conditions of flow quantities as opposed to approximating the Dirac-delta function by grid-dependent functions in the original immersed boundary method. This gives the IIM its advantage to resolve boundaries more accurately. We present our recent numerical results of 3D flow simulation using the IIM and also discuss a method for constructing the singular force in the case of prescribed boundary motions. [Preview Abstract] |
Sunday, November 20, 2005 9:31AM - 9:44AM |
AG.00008: Predicting Pressure Fluctuations in Large Eddy Simulations Using the Immersed Boundary Method Seongwon Kang, Gianluca Iaccarino, Parviz Moin, Frank Ham An immersed boundary (IB) method for Large Eddy Simulations (LES) is developed and applied to the turbulent flow around an airfoil. The numerical procedure is based on the use of local mesh refinement to achieve high resolution in the turbulent boundary layer. The primary objective of the paper is to evaluate the accuracy of the wall-pressure spectra predictions. This is relevant for predicting the noise generated by the airfoil. The influence of the IB wall treatment is investigated by applying different reconstruction schemes; in particular, an interpolation method corrected to enforce global mass conservation is introduced and tested. The computed wall- pressure spectra as well as averaged flow fields are compared to experiment and to previous LES simulations carried out using a body-fitted mesh. [Preview Abstract] |
Sunday, November 20, 2005 9:44AM - 9:57AM |
AG.00009: An embedded boundary formulation for fluid structure interactions of elastically mounted rigid bodies Jianming Yang, Elias Balaras We will present an embedded-boundary formulation that is applicable to fluid structure interactions of elastically mounted rigid bodies. In this approach, the Navier-Stokes equations for incompressible flow are solved on fixed Cartesian grids with the immersed moving bodies treated using a second-order sharp- interface embedded boundary method. The ODE's governing the rigid body motions are solved using Hamming's fourth-order predictor-corrector method. A strong coupling scheme is adopted, where the fluid and the structure are treated as elements of a single dynamical system, and all of the governing equations are integrated simultaneously, and interactively in the time-domain. Vortex-induced vibrations of a circular cylinder with one and two degrees-of-freedom are studied and the results are in good agreement with reference experiments and simulations. The interactions of multiple elastic cylinders have also been simulated to demonstrate the robustness of the proposed approach in cases involving multiple bodies. [Preview Abstract] |
Sunday, November 20, 2005 9:57AM - 10:10AM |
AG.00010: Numerical Simulation of Thermal Process in an Industrial Rotary Furnace Zeyi Jiang, Xiaobing Yang, Xinxin Zhang A numerical simulation was performed for the complex thermal processes of heating steel bars in a rotary furnace, which involve both the momentum transfer and the energy transfer mainly by radiation and combustion. A CFD commercial software CFX was employed to solve the proposed 2-D mathematical model. The boundary conditions for the simulation were initially chosen basing on on-line measured data of the products. The temperature and velocity profiles inside the furnace will be obtained as well as the other related thermal parameters. By adjusting the boundary conditions, the proposed model can be used for the dynamical control of the furnace. Meanwhile, using the radiative heat flux from the furnace wall, the surface temperature of steel bar can be obtained from the simulation. The total heat exchange factor of each control zone in the furnace is analyzed along the circumferential and axial directions of the steel bar, as well as among different zones of the furnace. These total heat exchange factors will be further used to improve the model for simplifying the online control. [Preview Abstract] |
Sunday, November 20, 2005 10:10AM - 10:23AM |
AG.00011: A New High-Order Immersed Interface Method for Numerical Simulation of Two-Phase Flow Xiaolin Zhong In recent years, there has been a strong interest in developing robust and accurate numerical methods for simulating complex flow with imbedded interface of discontinuity. Numerical simulations for these flows are challenging in many aspects. First, the interface geometry can be complex and can undergo change, merge and breakup during the course of the simulation. The most popular methods in modeling flow with interface surfaces are, among others, the volume of fluid method, front tracking method, and level-set methods. Second, the flow variables and their derivatives are usually not continuous across the interface. This discontinuity poses severe limitation in the accuracy of common CFD methods. The current available methods in treating the interface jump conditions, such as the Immerse Boundary Method, are mostly first order accurate at the interface. The Immersed Interface Method of Laveque and Li (1994) can reach a second order accuracy. But this method is very difficult to apply to complex 3-D flow problems. We present a new general high-order immersed interface method (2nd and 4th order in particular) for two-phase flow simulation. The method can be of arbitrarily high-order accuracy by using additional grid points on both sides of the interface, instead of jump conditions of higher order derivatives. The finite difference formulas for irregular grid points near the interface are derived either by a one-sided Taylor expansion or by a matched polynomial interpolation. [Preview Abstract] |
Sunday, November 20, 2005 10:23AM - 10:36AM |
AG.00012: Modeling fluid flow and heat transfer in PEM fuel cell using lattice Boltzmann approach Behnam Afsharpoya, Lian-Ping Wang The fluid flow and species transport in fuel cells are affected by diffusion, advection, thermal gradients, material properties, electrochemical effects, and interfacial forces. A consistent approach capable of modeling these processes has not yet been developed. There have been studies addressing transport of reactants and products in the gas phase, however, water management and convective / thermal effects are still poorly understood. While most modeling efforts in fuel-cell research adopt the traditional CFD approach based on the continuum governing equations, we are developing lattice- Boltzmann (LB) methods to model fluid and thermal transport inside flow channels and gas diffusion layers in proton exchange membrane fuel cells. Specifically, we have developed and tested a new method for implementing structured non-uniform mesh using Lagrangian interpolations. A three-dimensional LB code has been developed for thermal flows through a section of serpentine channel with a gas diffusion layer. The gas diffusion layer is modelled as a porous medium using a modified LB equation and a forcing term. A separate distribution is used to model thermal effects. Methods of validating the approach and preliminary results will be presented. [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