Bulletin of the American Physical Society
72nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 64, Number 13
Saturday–Tuesday, November 23–26, 2019; Seattle, Washington
Session S41: CFD: General II |
Hide Abstracts |
Chair: Omar Matar, Imperial College London Room: 6c |
Tuesday, November 26, 2019 10:31AM - 10:44AM |
S41.00001: Surrogate-model Optimisation Strategies for non-Newtonian Gelling Adam Kutnar, Lachlan Mason, Indranil Pan, Richard Craster, Omar Matar Highly resolved simulations have advanced to the extent that they are routinely relied upon in engineering design. The trade-off for this accuracy, however, can be prohibitively high computational cost. These costs are further compounded when multiple calls to an expensive simulation code are required, for example when determining the optimal parameters for maximising an engineering performance metric. Here, simple strategies such as grid searching do not scale well with the number of optimisation parameter dimensions. In this study, we investigate industrial gelling phenomena for non-Newtonian fluids in eccentric annuli. The task is to minimise gelling under engineering constraints on both flow and geometric properties. We achieve computational tractability via a surrogate model optimisation strategy and demonstrate its benefits to industrial simulation practitioners, including a reduction in design lead times, by benchmarking against conventional grid and random sampling methods. Best practices are highlighted for a class of generic fluid dynamics optimisation problems [Preview Abstract] |
Tuesday, November 26, 2019 10:44AM - 10:57AM |
S41.00002: Geometric parameter optimization of a liquid jet liquid ejector Ivan Korkischko, Felipe Silva Maffei, Rafael dos Santos Gioria, Julio Romano Meneghini Ejectors are devices employed as pumps or compressors, which work transferring momentum from a primary fluid (high pressure) to a secondary fluid (low pressure). On the one hand, their main advantages over standard pumps and compressors are no moving parts, no need of lubricants and seals, and low noise and maintenance. On the other hand, ejectors have low efficiency compared to other devices and a very narrow region of optimal operation. Thus, ejectors certainly benefit from optimization studies. This investigation was based on a CFD model of a liquid jet liquid (LJL) ejector. The finite element method was used, coupled with the k-epsilon turbulence model. The optimization study had three steps. First, the constants of the turbulence model were recalibrated to minimize the difference between the numerical and experimental efficiency curves. Second, using the main geometric parameters as control variables, the peak efficiency was maximized. Finally, the optimized geometry was further improved, considering the transitions between the different ejector components, which were originally sharp corners. The optimized round corners increased the ejector efficiency. [Preview Abstract] |
Tuesday, November 26, 2019 10:57AM - 11:10AM |
S41.00003: Direct Numerical Simulations of Mixing: From Aeration Tanks to Food Mixers Seungwon Shin, Lyes Kahouadji, Jalel Chergui, Damir Juric, Richard Craster, Omar Matar The dynamics of stirred tanks has been studied over a wide range of laboratory and industrially conditions and scales, in order to improve the mixing efficiency. We use a three-dimensional two-phase flow dynamics solver coupled with a direct Forcing Method to handle accurately the fluid structure interaction occurring in any types of stirred tank. The numerical framework employed here circumvents numerous meshing issues normally associated with constructing complex geometries (impellers, baffles, etc.) within typical computational fluid dynamics packages. All these solid structures are constructed via a module that defines solid objects by means of a static distance function. Typical examples will be presented such us aeration (bubbly mixing), cavern formation (stirring viscoplastic fluids), and typical “Egg Beaters” with different types of fluid rheology. [Preview Abstract] |
Tuesday, November 26, 2019 11:10AM - 11:23AM |
S41.00004: Numerical Simulation of Crude-Oil Fouling with the Volume-of-Fluid Method Gabriel Goncalves, Mirco Magnini, Omar Matar In oil production or processing pipelines, the change in themodynamical conditions may cause components of crude-oil to precipitate and adhere to the wall, leading to undesired changes in the hydaulic and thermal performance of the system over time. Although significant progress has been achieved in understanding qualitatively the main mechanisms of wax formation and removal, quantitative comparisons with experimental data are still heavily dependent on calibration to experimental data. In order to perform high-fidelity simulations in arbitrary geometries, a two-phase multi-component solver with heat transfer and phase change was implemented in the OpenFOAM open-source framework. The volume-of-fluid method is used for tracking the interface between fouling layer and liquid phase. The implementation was compared with previous calculations performed in a commercial CFD platform and preliminarily validated with experimental data from the literature. [Preview Abstract] |
Tuesday, November 26, 2019 11:23AM - 11:36AM |
S41.00005: Three-dimensional Numerical Simulations of Turbulent Surfactant-laden Jets Jalel Chergui, Ricardo Constane-Amores, Assen Batchvarov, Lyes Kahouadji, Seungwon Shin, Damir Juric, Richard Craster, Omar Matar Liquid atomisation processes are widely used to break down a liquid stream into smaller droplets to enhance its mixing with a stagnant phase. These streams may be contaminated with surfactants, whose concentration variations lead to gradients in surface tension and associated Marangoni stresses. Here, we study, for the first, the effect of surfactant on the complex interfacial dynamics associated with a turbulent jet. We use a hybrid front-tracking/level-set method to capture the dynamics of the complex topological changes in this flow. The numerical method allows the natural tracking of the concentration of interfacial surfactant species and the faithful modelling of its spatio-temporal evolution. Our model also accounts for surfactant solubility and bulk-interface mass exchange. We perform a full parametric study of the effect of surfactant properties on the dynamics. The effect of Marangoni stresses is analysed in terms of the mechanisms giving rise to the droplet size distributions depending on the elasticity number. An attempt to understand the interaction between the observed vortical structures accompanying the flow and the regions of elevated surfactant concentration will also be presented. [Preview Abstract] |
Tuesday, November 26, 2019 11:36AM - 11:49AM |
S41.00006: Computational Investigation of the Retropulsive Jet Produced by Antral Contraction Waves in a Model Stomach Kathleen Feigl, Samer Alokaily, Franz Tanner Numerical simulations are performed to investigate the retropulsive jet that is produced by peristaltic motion during the mixing and digestion process in a model human stomach. The geometrical model for the stomach consists of an axisymmetric conical-shaped tube with a wall at one end which represents the antrum and closed pylorus. The antral contraction waves which produce the peristaltic flow, and consequently the retropulsive jet near the closed pylorus, are modeled as traveling waves along the tube boundary which deform the computational mesh. This geometrical model and the boundary deformation algorithm are implemented into a C++ library and then coupled with the open source code OpenFOAM. The effect of various parameters on the retropulsive jet near the pylorus is investigated. These parameters include the fluid viscosity, wave speed, wave width and maximum relative occlusion. The retropulsive jet is quantified by its peak velocity and jet length along the centerline at maximum relative occlusion. For each wave geometry, it is found that the velocity and pressure curves scale with wave speed at low Reynolds numbers. Between different wave geometries, scaling laws are proposed and tested for the peak centerline velocity and jet length at low Reynolds numbers. [Preview Abstract] |
Tuesday, November 26, 2019 11:49AM - 12:02PM |
S41.00007: Hydrodynamics of solid objects impacting on free surface fluid Deepak Kumar Pandey, HeeChang Lim The impact studies of different shape and size of solid objects and curved plates on the free surface of fluid have a great practical significance from engineering perspectives. It is worth noting here that such type of impact phenomena can be correlated with slamming of ship bow, seaplanes, submarines, military projectiles, etc. in the seawater. The objectives of this paper are to experimentally and numerically analyze the field variables (pressure distribution, splashing, wave propagation, etc.) arising out of free fall of different shape and size of solid objects (square prism, rectangular prism, triangular pyramid, cylinder, sphere, etc.) on the free surface of the fluid. The fluids employed were water and oils having different viscosity. During the experiments, free-falling solid objects as projectiles were allowed to impact the free surface of the fluid. The height and the angle of the free falling solid objects were also varied in order to observe its effect on splashing and wave propagation. The interface behaviors and wave propagation have been captured using a high-speed camera. However, pressure sensors and strain gauges were employed in order to record the pressure and impact load at different locations. The simulation results validate the experimental results. [Preview Abstract] |
Tuesday, November 26, 2019 12:02PM - 12:15PM |
S41.00008: Active learning methodologies for surrogate model development in CFD applications Indranil Pan, Gabriel Goncalves, Assen Batchvarov, Yuxin Liu, Yuyi Liu, Vikneswaran Sathasivam, Nicholas Yiakoumi, Lachlan Mason, Omar Matar Computational fluid dynamic simulations typically have high computational costs, such that for parametric analysis and engineering design an inexpensive surrogate model, which is capable of reproducing the trends of some variables of interest, may be desired. However, generating regressions based on a full grid-based parametric variation is generally infeasible even for a system with moderate number of parameters. In this work, a wide array of active learning techniques were coupled with different regression models to achieve high predictive performance under the constraints of a limited function evaluation budget. The case studies involve flows of industrial relevance and the results outline some best practices for such simulations and highlights future research directions. [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