Bulletin of the American Physical Society
71st Annual Meeting of the APS Division of Fluid Dynamics
Volume 63, Number 13
Sunday–Tuesday, November 18–20, 2018; Atlanta, Georgia
Session A09: Computational Fluid Dynamics Methods for Multiphase Flows I |
Hide Abstracts |
Chair: Sivaramakrishnan Balachandar, University of Florida Room: Georgia World Congress Center B214 |
Sunday, November 18, 2018 8:00AM - 8:13AM |
A09.00001: Modeling approaches for bubbly, cavitating flows Spencer H. Bryngelson, Tim Colonius We compare two approaches for modeling bubbly, cavitating flows. The first approach is a deterministic model, for which bubbles are represented in a Lagrangian framework as advected particles, each sampled from a distribution of equilibrium bubble sizes. The dynamic coupling to the liquid phase is modeled through local volume averaging. The second approach is stochastic; ensemble-phase averaging is used to derive mixture-averaged equations and the field equations are evolved in an Eulerian reference frame for the associated bubble properties, each representing bins of an underlying equilibrium distribution. In both cases the equations are closed by solving Rayleigh-Plesset-like equations for the bubble dynamics as forced by the local or mixture-averaged pressure, respectively. Computationally, there are complex tradeoffs between these two approaches, especially for modern, parallel architectures. As a step towards negotiating this landscape, we simulate an acoustically excited dilute bubble screen and compare the computationally efficiency of the two approaches, testing the sensitivity of each to resolution and under-sampling. |
Sunday, November 18, 2018 8:13AM - 8:26AM |
A09.00002: A new implicit finite element model for the analysis of droplet dynamics Alex Jarauta, Pavel Ryzhakov, Jordi Pons-Prats, Marc Secanell Two-phase flow at the sub-millimeter scale can be found in relevant industrial applications, such as inkjet printing devices or fuel cells. Droplet dynamics modeling can be a challenging task due to the reduced spatial and time scales involved in the acting capillary forces. The majority of two-phase models in literature use a fixed grid (Eulerian) framework for the description of the continuum. These methods may either have mass conservation issues or lead to artificial interface diffusion unless special precautions are taken. In addition, explicit integration of the surface tension term used in the majority of existing models leads to a capillary time step constraint. A new implicit surface tension model has been developed to analyze droplet dynamics. A Lagrangian framework is chosen to accurately track the domain boundary, which is critical for the introduction of the surface tension term for microfluidic applications. Different numerical benchmark examples, such as static, dynamic and sessile droplet, and capillary wave examples have been considered. The model is first order in time, and second order of convergence in space. Comparison with other models in literature shows that the model is stable for time steps as high as 20 times larger than previously reported values. |
Sunday, November 18, 2018 8:26AM - 8:39AM |
A09.00003: Discontinuous Galerkin simulation of rarefied dusty gas flows based on a second-order Boltzmann continuum model Rho Shin Myong, Omid Ejtehadi, Amin Rahimi The rocket plume impingement on the lunar surface can cause significant dust dispersal when the lunar lander approaches the landing site. In this work, we aim to study the rocket plume-lunar surface interaction and subsequent regolith particle dispersal within the Eulerian framework. The subjects are challenging due to the facts that, first, the conventional NSF equations may not be valid in rarefied dusty gas flows; second, the flow physics becomes inherently multi-physics, multi-scale problem in which the erosion of regolith, the interaction of gas and solid particle, and the dispersal of dust are coupled; and lastly, it is impossible to reproduce the lunar condition—characterized by micro-gravity, near-vacuum, and unique properties of the regolith—for an experimental test on the earth. Here, using a modal explicit discontinuous Galerkin method of a second-order Boltzmann continuum model that can cover both the continuum and free-molecular flows with a single framework, inherent complexities consisting of various flow regimes—the plume expanding in vacuum, shock and stagnation regions, local erosion, supersonic dusty jet flow, rarefied flow—are investigated in detail. |
Sunday, November 18, 2018 8:39AM - 8:52AM |
A09.00004: Eulerian-Lagrangian simulation of non-isothermal particle-laden flow Farid Rousta, Bamdad Lessani Direct numerical simulations of particle-laden channel flow with walls kept at different temperatures are performed. The position, velocity and temperature of each particle are computed in time with a Lagrangian approach, with a two-way coupling between particles and carrier flow in the heat and momentum equations. The focus of this study is on the effect of direct particle-to-wall heat conduction for wall-colliding particles. A model based on Hertzian contact is implemented in the in-house Eulerian-Lagrangian solver. It is shown that for smooth walls (with a very thin layer of fluid between the particles and wall during collision), and metallic particles, the particle-to-wall heat conduction is not negligible, and it can even be comparable with the convective fluid-to-wall heat transfer. The results are presented in terms of different average wall roughness and particle mass loadings. |
Sunday, November 18, 2018 8:52AM - 9:05AM |
A09.00005: Embedded semi-analytical modeling of small scales in simulations of multiphase flow Alberto RomanAfanador, Gretar Tryggvason, Jiacai Lu Multiphase flows are notorious for generating ``features,’’ such as drops, threads and films, much smaller than the dominant interface scales. For small enough scales, surface tension and viscous effects are usually strong so the geometry and the flow are relatively simple and can be described by semi-analytical models. Resolving the smallest scales fully in numerical simulations is expensive and we explore how semi-analytical models can be coupled with fully resolved simulations of the rest of the flow, to reduce the resolution requirement. Small-scale structures are ``collapsed’’ to singularities (points, lines and sheets) by applying filtering to the interface and equations for the evolution of the singularity strength and their coupling with the resolved fluid are derived. The singularities are captured using Lagrangian points advected by the flow, while the rest of the governing equations are solved using a Eulerian approach. The idea and the approach are described and results for simple test cases, as well as preliminary results for a 2D perturbed jet are shown. The results are evaluated by comparison with fully resolved simulations. Low dimensional models such as this one can potentially reduce the computational cost of multiscale and multiphase simulations. |
Sunday, November 18, 2018 9:05AM - 9:18AM |
A09.00006: A novel Euler-Lagrange method that incorporates fully resolved physics using pairwise interaction extended point-particle (PIEP) model. S. Balachandar, W. C. Moore, Georges Akiki, Kai Liu In the traditional Euler-Lagrange (EL) method the point-particle (PP) force models are based on flow Reynolds number and local particle volume fraction as seen by the particle. The presence of neighboring particles is only taken into account on average through the local particle volume fraction. Here we start with the complete N-particle problem and employ pairwise-interaction and order-invariance assumptions to rigorously simplify the problem and develop a new framework of pairwise interaction extended point-particle (PIEP) approach for coupling the particulate and fluid phases. The key ingredient in this approach is axisymmetric maps of perturbation effects of a neighbor, which can be thought of as richer drag/lift laws. We will discuss how machine learning can be used, guided by physical understanding, to develop these maps from the DNS data of flow over a random array of particles. |
Sunday, November 18, 2018 9:18AM - 9:31AM |
A09.00007: Convergence of Point Particle Models in Euler-Lagrange Simulations of Shock-Particle Interaction Samaun Nili, Chanyoung Park, Raphael T. Haftka, Nam H. Kim, S. Balachandar Point particle methods are extensively used in simulating Euler-Lagrange multiphase dispersed flow. However, numerical convergence and accuracy of these methods under mesh refinement is still an open question. The standard approach of approximating the fluid-particle coupling at the particle center fails to converge as the Eulerian grid is reduced below particle size. To quantify the spatial discretization error, we consider the Faxèn form of the coupling between the particle and the fluid to account for the finite size of the particle and demonstrate this approach permits convergence. The generalized Faxén form also allows for apportioning of the different force components to fluid cells based on the fraction of particle volume or the fraction of particle surface area in the cell. We tested this approach on a four-way coupled one-dimensional model of shock propagation through a particle-laden field at moderate volume fraction, where the convergence is achieved for a well-formulated force model and back coupling for finite size particles. |
Sunday, November 18, 2018 9:31AM - 9:44AM |
A09.00008: Analysis of particle wakes for PIEP modeling in Euler-Lagrange simulations W. C. Moore, S. Balachandar To aid in the understanding and modeling of microscale physics, particle resolved direct numerical simulations have become more prevalent in recent years. From these simulations, models can be developed to approximate the microscale inter-phase forces, Reynolds stresses, and residual stresses. The typical approach is to model these properties as functions of the macroscale Reynolds number and particle volume fraction, accounting for the average effect of microscale contributions. Instead, this study accounts for the effects of each neighbor directly. We define a linearly superposable wake (LSW), which when added for all the particles best captures microscale properties of a DNS. The LSW is then used to model the inter-phase forces, Reynolds stresses, and residual stresses in the pairwise interaction expended point-particle (PIEP) framework. This particle based view allows for more precise approximations. In the dilute volume fraction limit, the inter-phase force model approaches the PIEP model based on an isolated particle wake. However, at higher volume fractions, LSW is now calculated using regression applied to DNS data. |
Sunday, November 18, 2018 9:44AM - 9:57AM |
A09.00009: Applications of pairwise interaction extended point-particle model: mid-field spray control Kai Liu, S Balachandar This study applies the pairwise interaction extended point-particle (PIEP) model to accurately predict the hydrodynamic influence between neighboring droplets dispersing in the externally controlled mid-field jet spray in an Euler-Lagrangian simulation. The two-way coupled Euler-Lagrangian methodology has been widely implemented in spray simulations, since it allows efficient tracking of millions of droplets without resolving the micro-scale flow around them. However, the accuracy of this approach suffers from the neglect of neighboring droplets interaction. In this talk we will first illustrate the importance of neighbor-neighbor interaction in the context of a sedimenting particle system. We will then consider simulations of mid-field evolution of a droplet spray with and without the PIEP model and evaluate the importance of accounting for droplet-droplet interaction. Finally we will consider the possibility of controlling the spray with acoustic forcing. |
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