Bulletin of the American Physical Society
68th Annual Meeting of the APS Division of Fluid Dynamics
Volume 60, Number 21
Sunday–Tuesday, November 22–24, 2015; Boston, Massachusetts
Session D9: CFD: Computational Methods and Modeling of Multiphase Flows I |
Hide Abstracts |
Chair: Florian Kummer, Technical Univiversity of Darmstadt Room: 109 |
Sunday, November 22, 2015 2:10PM - 2:23PM |
D9.00001: Taylor bubbles at high viscosity ratios: experiments and numerical simulations Buddhika Hewakandamby, Abbas Hasan, Barry Azzopardi, Zhihua Xie, Chris Pain, Omar Matar The Taylor bubble is a single long bubble which nearly fills the entire cross section of a liquid-filled circular tube, often occurring in gas-liquid slug flows in many industrial applications, particularly oil and gas production. The objective of this study is to investigate the fluid dynamics of three-dimensional Taylor bubble rising in highly viscous silicone oil in a vertical pipe. An adaptive unstructured mesh modelling framework is adopted here which can modify and adapt anisotropic unstructured meshes to better represent the underlying physics of bubble rising and reduce computational effort without sacrificing accuracy. The numerical framework consists of a mixed control volume and finite element formulation, a ‘volume of fluid’-type method for the interface-capturing based on a compressive control volume advection method, and a force-balanced algorithm for the surface tension implementation. Experimental results for the Taylor bubble shape and rise velocity are presented, together with numerical results for the dynamics of the bubbles. A comparison of the simulation predictions with experimental data available in the literature is also presented to demonstrate the capabilities of our numerical method. [Preview Abstract] |
Sunday, November 22, 2015 2:23PM - 2:36PM |
D9.00002: CFD-informed unified closure relation for the rise velocity of Taylor bubbles in pipes Enrique Lizarraga-Garcia, Jacopo Buongiorno, Eissa Al-Safran, Djamel Lakehal Two-phase slug flow commonly occurs in gas and oil systems. Current predictive methods are based on the mechanistic models, which require the use of closure relations to complement the conservation equations to predict integral flow parameters such as liquid holdup (or void fraction) and pressure gradient. Taylor bubble velocity in slug flow is one of these closure relations which has been determined to significantly affect the calculation of these parameters. In this work, Computational Fluid Dynamics (CFD) with Level-Set as the Interface Tracking Method (ITM) are employed to simulate the motion of Taylor bubbles in slug flow, for which the commercial code TransAT is used. A large numerical database with stagnant and flowing liquid for various Reynolds numbers is being generated from which a unified Taylor bubble velocity correlation in stagnant liquids for an ample range of fluid properties and pipe geometries is proposed ($Mo\in [1\cdot10^{-6},5\cdot10^3], Eo\in [10,700]$). Furthermore, it is found that the velocity of Taylor bubbles in inclined pipes is greatly affected by the presence of a lubricating thin film between the bubble and the pipe wall. An analytical and experimentally validated criterion, which predicts the film existence, draiage and breakup, is presented. [Preview Abstract] |
Sunday, November 22, 2015 2:36PM - 2:49PM |
D9.00003: Detached eddy simulations of Taylor bubbles rising in stagnant liquid columns Hassan Shaban, Stavros Tavoularis The rise of a single air Taylor bubble in a vertical circular tube filled with stagnant water was investigated numerically using the Volume Of Fluid (VOF) method to model the phase distribution and the Detached Eddy Simulation (DES) method for turbulence modelling. The predictions were in good quantitative agreement with previous experimental results. The simulation results provided insight into bubble shedding in the wake of the Taylor bubble, frictional pressure drop along the tube and scalar dispersion caused by the passage of the Taylor bubble. The interaction between adjacent Taylor bubbles and the process of Taylor bubble coalescence were also examined in detail. [Preview Abstract] |
Sunday, November 22, 2015 2:49PM - 3:02PM |
D9.00004: Numerical modeling of turbulent swirling flow in a multi-inlet vortex nanoprecipitation reactor using dynamic DDES James C. Hill, Zhenping Liu, Rodney O. Fox, Alberto Passalacqua, Michael G. Olsen The multi-inlet vortex reactor (MIVR) has been developed to provide a platform for rapid mixing in the application of flash nanoprecipitation (FNP) for manufacturing functional nanoparticles. Unfortunately, commonly used RANS methods are unable to accurately model this complex swirling flow. Large eddy simulations have also been problematic, as expensive fine grids to accurately model the flow are required. These dilemmas led to the strategy of applying a Delayed Detached Eddy Simulation (DDES) method to the vortex reactor. In the current work, the turbulent swirling flow inside a scaled-up MIVR has been investigated by using a dynamic DDES model. In the DDES model, the eddy viscosity has a form similar to the Smagorinsky sub-grid viscosity in LES and allows the implementation of a dynamic procedure to determine its coefficient. The complex recirculating back flow near the reactor center has been successfully captured by using this dynamic DDES model. Moreover, the simulation results are found to agree with experimental data for mean velocity and Reynolds stresses. [Preview Abstract] |
Sunday, November 22, 2015 3:02PM - 3:15PM |
D9.00005: Development of multiphase Navier-Stokes simulation capability for turbulent gas flow over laminar liquid for Cartesian grids Sha Miao, Kelli Hendrickson, Yuming Liu, Hariprasad Subramani This work presents a novel and efficient Cartesian-grid based simulation capability for the study of an incompressible, turbulent gas layer over a liquid flow with disparate Reynolds numbers in two phases. This capability couples a turbulent gas-flow solver and a liquid-layer based on a second-order accurate Boundary Data Immersion Method (BDIM) at the deformable interface. The turbulent gas flow solver solves the incompressible Navier-Stokes equations via direct numerical simulation or through turbulence closure (unsteady Reynolds-Averaged Navier-Stokes Models) for Reynolds numbers O($10^6$). In this application, a laminar liquid layer solution is obtained from depth-integrated Navier-Stokes equations utilizing shallow water wave assumptions. The immersed boundary method (BDIM) enforces the coupling at the deformable interface, the boundary conditions to turbulence closure equations and defines the domain geometry on the Cartesian grid. Validations are made for the turbulent gas channel flow over high-viscosity liquid. This simulation capability can be applied to problems in the oil and industrial sector such as channel and pipe flows with heavy oils as well as wind wave generation in shallow waters. [Preview Abstract] |
Sunday, November 22, 2015 3:15PM - 3:28PM |
D9.00006: New techniques for meshless flow simulation generalizing moving least squares Nathaniel Trask, Martin Maxey While the Lagrangian nature of SPH offers unique flexibility in application problems, practitioners are forced to choose between compatibility in div/grad operators or low accuracy limiting the scope of the method. In this work, two new discretization frameworks are introduced that extend concepts from finite difference methods to a meshless context: one generalizing the high-order convergence of compact finite differences and another generalizing the enhanced stability of staggered marker-and-cell schemes. The discretizations are based on a novel polynomial reconstruction process that allows arbitrary order polynomial accuracy for both the differential operators and general boundary conditions while maintaining stability and computational efficiency. We demonstrate how the method fits neatly into the ISPH framework and offers a new degree of fidelity and accuracy in Lagrangian particle methods. [Preview Abstract] |
Sunday, November 22, 2015 3:28PM - 3:41PM |
D9.00007: Methods to Prescribe Particle Motion to Minimize Quadrature Error in Meshfree Methods Jeremy Templeton, Lindsay Erickson, Karla Morris, David Poliakoff Meshfree methods are an attractive approach for simulating material systems undergoing large-scale deformation, such as spray break up, free surface flows, and droplets. Particles, which can be easily moved, are used as nodes and/or quadrature points rather than a relying on a fixed mesh. Most methods move particles according to the local fluid velocity that allows for the convection terms in the Navier-Stokes equations to be easily accounted for. However, this is a trade-off against numerical accuracy as the flow can often move particles to configurations with high quadrature error, and artificial compressibility is often required to prevent particles from forming undesirable regions of high and low concentrations. In this work, we consider the other side of the trade-off: moving particles based on reducing numerical error. Methods derived from molecular dynamics show that particles can be moved to minimize a surrogate for the solution error, resulting in substantially more accurate simulations at a fixed cost. Sandia National Laboratories is a multiprogram laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. [Preview Abstract] |
Sunday, November 22, 2015 3:41PM - 3:54PM |
D9.00008: A balanced-force conservative volume-of-fluid method for simulating two-phase flows on unstructured grids Christopher Ivey, Parviz Moin A balanced-forced conservative volume-of-fluid method for simulating two-phase flows on unstructured grids is presented. The two-phase Navier-Stokes equations are solved using a median-dual-partitioned collocated node-centered finite-volume discretization and a specialized fractional-step method. Conservative mass and momentum convection fluxes are calculated using a novel volume-of-fluid method. Accurate interface normals and curvatures are calculated on the non-convex median-dual mesh using the recently proposed embedded height-function technique. Spurious currents are minimized using a balanced-force algorithm and the continuum-surface force description of surface tension. The results of several two- and three-dimensional benchmark test cases on various unstructured meshes demonstrate the effectiveness of the proposed proposed two-phase flow solver. [Preview Abstract] |
Sunday, November 22, 2015 3:54PM - 4:07PM |
D9.00009: High-Order Discontinuous Galerkin Level Set Method for Interface Tracking and Re-Distancing on Unstructured Meshes Patrick Greene, Robert Nourgaliev, Sam Schofield A new sharp high-order interface tracking method for multi-material flow problems on unstructured meshes is presented. The method combines the marker-tracking algorithm with a discontinuous Galerkin (DG) level set method to implicitly track interfaces. DG projection is used to provide a mapping from the Lagrangian marker field to the Eulerian level set field. For the level set re-distancing, we developed a novel marching method that takes advantage of the unique features of the DG representation of the level set. The method efficiently marches outward from the zero level set with values in the new cells being computed solely from cell neighbors. Results are presented for a number of different interface geometries including ones with sharp corners and multiple hierarchical level sets. The method can robustly handle the level set discontinuities without explicit utilization of solution limiters. Results show that the expected high order (3rd and higher) of convergence for the DG representation of the level set is obtained for smooth solutions on unstructured meshes. High-order re-distancing on irregular meshes is a must for applications were the interfacial curvature is important for underlying physics, such as surface tension, wetting and detonation shock dynamics. [Preview Abstract] |
Sunday, November 22, 2015 4:07PM - 4:20PM |
D9.00010: High-order accurate multi-phase simulations: building blocks and whats tricky about them Florian Kummer We are going to present a high-order numerical method for multi-phase flow problems, which employs a sharp interface representation by a level-set and an extended discontinuous Galerkin (XDG) discretization for the flow properties. The shape of the XDG basis functions is dynamically adapted to the position of the fluid interface, so that the spatial approximation space can represent jumps in pressure and kinks in velocity accurately. By this approach, the `$h^p$-convergence' property of the classical discontinuous Galerkin (DG) method can be preserved for the low-regularity, discontinuous solutions, such as those appearing in multi-phase flows. Within the past years, several building blocks of such a method were presented: this includes numerical integration on cut-cells, the spatial discretization by the XDG method, precise evaluation of curvature and level-set algorithms tailored to the special requirements of XDG-methods. The presentation covers a short review on these building-block and their integration into a full multi-phase solver. A special emphasis is put on the discussion of the several pitfalls one may expire in the formulation of such a solver. [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