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 P02: Multiphase Flows: Modeling and Theory III 
Hide Abstracts 
Chair: Andrea Prosperetti, University of Houston Room: 2B 
Monday, November 25, 2019 5:16PM  5:29PM 
P02.00001: Interface Retaining Coarsening for Multiphase Flows Xianyang Chen, Jiacai Lu, Gretar Tryggvason Multiphase flows are characterized by sharp moving phase boundaries, separating different fluids or phases. In many cases the dynamics of the interfaces determines the behavior of the flow. In a coarse, or reduced order model, either an averaged twofluid model or a largeeddysimulation like one, it is therefore critical to retain a sharp interface for the resolved scales. The point particle model for disperse flows is a widely used limiting case. Different strategies to retain sharp interfaces are possible. In the simplest case the indicator function identifying the different fluids is filtered and the sharp interface restored by nonlinear postprocessing, consisting either of identifying the interface location from the filtered field, or let the interface ``flow’’ to its new location. Another approach is to work directly with Lagrangian marker points identifying the interface and average their coordinates, or evolve the interface based on curvature or other measures. The different approaches are discussed and compared, the relationship with image processing in computer graphics pointed out, and implications for the flow field are studied. Modeling approaches for the unresolved scales are briefly reviewed. [Preview Abstract] 

P02.00002: ABSTRACT WITHDRAWN 
Monday, November 25, 2019 5:42PM  5:55PM 
P02.00003: CFDPBM simulation of bubble columns based on fixed pivot method: Influence of the moment closure Xiaopeng Shang, Bing Feng Ng, Man Pun Wan, Shirun Ding The fixed pivot method can conserve two moments while other moments suffer from inherent errors caused by internal inconsistency. The influence of moment closure in the CFDPBM simulation is studied on population balance and hydrodynamics of bubble columns. The CFDPBM model, which conserves the surface area and volume (second and third moment, i.e. CFDPBMSV model), and the number and volume (zeroth and third moment, i.e. CFDPBMNV model), is developed based on a twofluid model. A rectangular bubble column is simulated by the CFDPBMSV and CFDPBMNV model, respectively. It is found that both models can capture the oscillating plume of the gasliquid flow inside the reactor. The flow features predicted by the CFDPBMSV model show better agreement with experimental data, in terms of the timeaveraged vertical liquid velocity, gas holdup and plume oscillation period, than the CFDPBMNV model. It is speculated that the better performance of the CFDPBMSV model is ascribed to more accurate predictions of interfacial forces and momentum transfer between two phases due to internal consistency of the local Sauter Mean Diameter compared to the CFDPBMNV model. [Preview Abstract] 
Monday, November 25, 2019 5:55PM  6:08PM 
P02.00004: TwoWay Coupled EulerLagrange Simulation on MidField Spray under MultiPhysics Control Kai Liu, Sivaramakrishnan Balachandar This study performs twoway coupled EulerLagrange simulations on the midfield liquidgas roundjet spray, externally controlled by multiphysics strategies. The twoway coupled EulerLagrange methodology has been commonly used in simulating complicated multiphase flows containing huge number of dispersed particles. To further improve the simulation accuracy, fluidmediated particleparticle interactions in subgrid scale is rigorously evaluated by the pairwise interaction extended pointparticle (PIEP) model. And the selfinduced velocity issue is also suppressed by the selfinduced correction (SIC) model. In this talk, we will first take sedimentation problem as an example to illustrate the effect of PIEP model and SIC model in optimizing EulerLagrange simulations. Then briefly valid the fundamental singlephase turbulent roundjet flow simulation according to experimental and theoretical statistics. Finally, EulerLagrange simulations incorporating these models will be demonstrated and compared with experiments. Hydrodynamic, electrostatic and ultrasonic controlling strategies will also be modelled to test their performance in changing the dispersion of spray droplets. [Preview Abstract] 
Monday, November 25, 2019 6:08PM  6:21PM 
P02.00005: Exploration of Gasliquid Flow Mixing Region for the Purpose of Driftflux Model Enhancement Ivan Nepomnyashchikh, James Liburdy The DriftFlux model (DFM) is a powerful reduce order model that can be applied to gasliquid flows. As applied to a one dimensional internal flow the model requires specification of two parameters: $C_{0}$ (distribution parameter) and $v_{s}$ (weighted mean drift velocity). It is the goal of the current study to develop a DriftFlux model appropriate for mixing regions in pipe flows for a range of gasliquid flow parameters that addresses both steady and transient flows. This study focuses on high fidelity simulations useful for the determination of both $C_{0}$ and $v_{s}$. A wide range of both transient and steady flow conditions is used (inlet parameters and fluid properties). The result is a range of two phase flow conditions, or flow map regimes, within the mixing region. For each set of conditions results are used to identify values of $C_{0}$ and $v_{s}$. Functional relationships are then found to map these parameters to the flow map regimes. The extent and range of the functional relationships allows for a robust set of conditions that can be utilized in the DriftFlux model. Using these results comparisons are made with existing functional forms of $C_{0}$ and $v_{s}$ to help identify the sensitivity to the transient conditions of mixing. [Preview Abstract] 
Monday, November 25, 2019 6:21PM  6:34PM 
P02.00006: Nearest Particle Statistics and Particlefluidparticle Stress of Multiphase Flows Duan Zhang This presentation starts with showing an important relation between the ensemble average and the average based on the nearest particle. Using this relation one can study longrange particlefluidparticle (PFP) interactions by defining an effective shortrange correlation accounting for effects of other surrounding particles. Physical meanings of this correlation will be presented, and the mathematical derivations will be outlined. This nearest particle average is compared with the average conditional on a pair of particles. For shortrange forces, the ensemble averages calculated from the both methods are the same. For longrange forces the nearest particle average has an advantage. Using the effective shortrange correlation force for hydrodynamic interactions among particles, a PFP stress is defined. The physical meanings of this stress will be examined. Using dilute potential and Stokes flows as examples, the PFP stresses are calculated. For cases of finite particle volume fractions, a numerical method is proposed to calculate this stress using the results of particle forces. This study of nearest particle statistics leads to many interesting and unanswered questions of particle interactions in disperse multiphase flows. [Preview Abstract] 
Monday, November 25, 2019 6:34PM  6:47PM 
P02.00007: A Physics/DataDriven Comprehensive Multiphase Force Coupling Model That Systematically Accounts For Clustering {\&} Shear Effects. Georges Akiki, Duan Zhang, S Balachandar Traditional approach to modeling the phase interaction has been through drag and lift force models that depend on Re and volume fraction $\theta $. However, it has been recognized that it is important to take into account local volume fraction gradients in particle distribution. Recently, it has been shown that the ParticleFluidParticle (PFP) stress can be rigorously defined, whose divergence accounts for the effect of inhomogeneous particle distribution. A series of fullyresolved DNS of flow around a uniform random distribution of particles are performed from which the PFP stress was evaluated. We then show how the DNS results can be accurately recovered with the pairwise interaction extended pointparticle (PIEP) model. This allows the evaluation of the PFP stress for a wide range of Re and $\theta $ at a negligible computational cost. We then test the PFP force in inhomogeneous distributions of particles by comparing the model prediction against results obtained from DNS. We also performed DNS of shear flow through a uniform random distribution of particles to establish the shearinduced lift force model at finite $\theta $. The significance of these new forces are then analyzed and compared to the mean streamwise force in uniform flows and to the mean streamwise and lateral forces in shear flows. [Preview Abstract] 
Monday, November 25, 2019 6:47PM  7:00PM 
P02.00008: Resolved Simulations of Particulate Shear Flow Gedi Zhou, Andrea Prosperetti This work describes the results of a numerical study of a suspension of thousands of resolved spheres in shear flow carried out by means of the Physalis method. The particle/fluid density ratio is varied between 2 and 5, the volume fraction between 5\% and 30\% and the particle Reynolds number between 25 and 50. Gravity is disregarded. The results shown include the particle mean free path, the pair distribution function, the particle diffusivity, collision frequency, the mixture stress and others. The mean free path decreases as the particle density is increased. Significant particle layering near the walls occurs with increasing volume fraction, with a strong effect on the velocity distribution of the suspension and the wall shear stress. The collisional contribution to the stress becomes dominant as the particle density ratio and volume fraction increase. [Preview Abstract] 
Monday, November 25, 2019 7:00PM  7:13PM 
P02.00009: Comprehensive Analysis of Two Fluid Models for Turbulent GasSolid and LiquidSolid Flow Omar Renzo Piminchumo Marinos, Donald J. Bergstrom In the Eulerian Two Fluid Model (TFM) formulation, the transport equation for the disperse phase consisting of solid particles is obtained from the Kinetic Theory of Granular Flow (KTGF). The transport equation for the carrier phase, either gas or liquid, closely resembles the singlephase Reynolds Averaged NavierStokes equation. Although successful predictions have been obtained for gassolid flow, simulations of liquidsolid flow typically show less agreement with the experimental data. The present work compares a TFM developed for gassolid flow with a recent TFM developed for liquidsolid flow: both describe the stresses of the solid phase using the KTGF. The main focus of the study is the difference in the predictions for the mean velocity, fluctuating velocity and volume fraction profiles, compared to experimental data for dilute liquidsolid flow in a vertical pipe. An outcome of the present research is a modification to the TFM which improves the predictions of the previous formulations. The modification considers changes to the model relations for the solid viscosity, granular temperature conductive coefficient and the source term intended to incorporate interstitial fluid effects. [Preview Abstract] 
Follow Us 
Engage
Become an APS Member 
My APS
Renew Membership 
Information for 
About APSThe American Physical Society (APS) is a nonprofit membership organization working to advance the knowledge of physics. 
© 2020 American Physical Society
 All rights reserved  Terms of Use
 Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 207403844
(301) 2093200
Editorial Office
1 Research Road, Ridge, NY 119612701
(631) 5914000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 200452001
(202) 6628700