Bulletin of the American Physical Society
70th Annual Meeting of the APS Division of Fluid Dynamics
Volume 62, Number 14
Sunday–Tuesday, November 19–21, 2017; Denver, Colorado
Session L36: Particle-Laden Flows: Turbulence IParticles Turbulence
|
Hide Abstracts |
Chair: Greg Voth, Wesleyan University Room: 302 |
Monday, November 20, 2017 4:05PM - 4:18PM |
L36.00001: Particle dynamics at the onset of shear Chung-min Lee, Armann Gylfason, Federico Toschi Small passive and inertial particles are present in many natural or industrial flows. The mixing and transportation of these particles are enhanced by turbulence. Many numerical and experimental studies have been carried out on the effect of turbulence on particle dynamics, but most of them deal with an underlying flow that is stationary. In many situations, however, a flow may evolve from being isotropic to anisotropic. Thus we aim to investigate particle dynamics during the development of a mean flow. We explore such flows by employing direct numerical simulation to simulate homogeneous isotropic turbulence and then apply a sudden shear deformation. Particles with various inertia are also simulated alongside of the flows with a couple of different shear rates. Lagrangian velocity and acceleration statistics and the changes with respect to the local flow field during the transition period will be reported. [Preview Abstract] |
Monday, November 20, 2017 4:18PM - 4:31PM |
L36.00002: Model based prediction of dynamics of particles in particle laden turbulent shear flow Partha Goswami, Swagnik Ghosh Particle-laden turbulent flows find application in wide range of industrial and natural processes. The advent of fast computing facility has enabled investigation of Particle-laden turbulent flows using Direct Numerical Simulation (DNS). Still simulating such flows in case of practically applicable geometry is still far from the reality. Therefore modeling such flows is inevitable. The proposed fluctuating force and fluctuating torque simulation is such a modeling method in which the effect of fluid velocity and vorticity fluctuations on the particle is modeled as anisotropic Gaussian white noise. For dilute suspensions, strength of the noise is extracted from diffusivity data of unladen fluid phase. The inter-particle and wall-particle collisions are modeled by introducing co-efficient of restitution ($e$) and roughness factor ($\beta$) in hard sphere collision model. Introduction of rotational diffusivity due to fluid vortical structures can predict the detailed rotational dynamics of particle phase. Present investigations have been performed for dilute sheared suspensions for different roughness factor in the limit of high Stokes number. The results obtained are compared with DNS using one-way coupling. [Preview Abstract] |
Monday, November 20, 2017 4:31PM - 4:44PM |
L36.00003: The role of particle-turbulence interactions on the pressure field near high-speed shear flows Jesse Capecelatro, Gregory Shallcross, David Buchta Heavy particles in turbulent flows, such as water droplets in air, are well-known to modify the carrier-phase velocity fluctuations. In high-speed flows, the turbulence provides a mechanism to radiate pressure fluctuations, which are usually considered in the safety and reliability of engineering applications, such as those environments near high-speed jets on aircraft carriers. In this presentation, we analyze the potential for reducing near-field pressure fluctuations via turbulence modulation by a disperse phase. Direct numerical simulations of particle-laden mixing layers are conducted for a range of Mach numbers, volume fractions, and Stokes numbers. Different turbulence regimes are identified based on the strength of interphase coupling characterized by the mass loading. The pressure intensity is observed to decrease with a comparable decrease in the turbulent kinetic energy. This reduction is found to be transient as the average volume fraction decreases with shear layer growth. In addition, we derive an evolution equation for the pressure variance in the presence of a disperse phase to quantify the particle-turbulence coupling mechanisms responsible for the observed reduction. [Preview Abstract] |
(Author Not Attending)
|
L36.00004: Capturing inertial particle transport in turbulent flows. Harry Stott, Andrew Lawrie, Robert Szalai The natural world is replete with examples of particle advection; mankind is both a beneficiary from and sufferer of the consequences. As such, the study of inertial particle dynamics, both aerosol and bubble, is vitally important. In many interesting examples such as cloud microphysics, sedimentation, or sewage transport, many millions of particles are advected in a relatively small volume of fluid. It is impossible to model these processes computationally and simulate every particle. Instead, we advect the probability density field of particle positions allowing unbiased sampling of particle behaviour across the domain. Given a 3-dimensional space discretised into cubes, we construct a transport operator that encodes the flow of particles through the faces of the cubes. By assuming that the dynamics of the particles lie close to an inertial manifold, it is possible to preserve the majority of the inertial properties of the particles between the time steps. We demonstrate the practical use of this method in a pair of instances: the first is an analogue to cloud microphysics- the turbulent breakdown of Taylor Green vortices; the second example is the case of a turbulent jet which has application both in sewage pipe outflow and pesticide spray dynamics. [Preview Abstract] |
Monday, November 20, 2017 4:57PM - 5:10PM |
L36.00005: Relative velocities in bidisperse turbulent suspensions Jan Meibohm We investigate the distribution of relative velocities between small heavy particles of different Stokes numbers in turbulence by analysing a statistical model for turbulent suspensions of particles with two different Stokes numbers. When the Stokes numbers are similar, the distribution exhibits power-law tails, just as in the case of equal Stokes numbers. We show that the power-law exponent is a non-analytic function of the mean $\overline{\text{St}}$ of the two Stokes numbers. This means that the exponent cannot be calculated in perturbation theory around the advective limit. When the difference between the Stokes numbers is larger, the power law disappears, but the tails of the distribution still dominate the relative-velocity moments, if $\overline{\text{St}}$ is large enough. [Preview Abstract] |
Monday, November 20, 2017 5:10PM - 5:23PM |
L36.00006: Anisotropic Stochastic Vortex Structure Method for Simulating Particle Collision in Turbulent Shear Flows Farzad Dizaji, Jeffrey Marshall, John Grant, Xing Jin Accounting for the effect of subgrid-scale turbulence on interacting particles remains a challenge when using Reynolds-Averaged Navier Stokes (RANS) or Large Eddy Simulation (LES) approaches for simulation of turbulent particulate flows. The standard stochastic Lagrangian method for introducing turbulence into particulate flow computations is not effective when the particles interact via collisions, contact electrification, etc., since this method is not intended to accurately model relative motion between particles. We have recently developed the stochastic vortex structure (SVS) method and demonstrated its use for accurate simulation of particle collision in homogeneous turbulence; the current work presents an extension of the SVS method to turbulent shear flows. The SVS method simulates subgrid-scale turbulence using a set of randomly-positioned, finite-length vortices to generate a synthetic fluctuating velocity field. It has been shown to accurately reproduce the turbulence inertial-range spectrum and the probability density functions for the velocity and acceleration fields. In order to extend SVS to turbulent shear flows, a new inversion method has been developed to orient the vortices in order to generate a specified Reynolds stress field. The extended SVS method is validated in the present study with comparison to direct numerical simulations for a planar turbulent jet flow. [Preview Abstract] |
Monday, November 20, 2017 5:23PM - 5:36PM |
L36.00007: Lagrangian structures in fluid turbulence damped by the addition of particles Yoichi Mito Lagrangian structures in turbulent gas flow being damped by the addition of a small amount of solid particles have been examined using a direct numerical simulation to calculate the gas velocities seen by particles and a point force method to calculate the forces exerted by particles on the gas. A simplified non-stationary flow model in which a uniform distribution of spherical particles are added into fully-developed turbulent gas flow through a vertical channel is considered. The Lagrangian autocorrelations of gas velocity fluctuations and of those seen by solid particles are calculated by setting point sources of fluid particles and of solid particles that do not exert forces on the gas at several distances from the wall and at several times after the addition of the solid particles that exert forces on the gas. Increases in the Lagrangian time scales for the gas velocity fluctuations and for those seen by solid particles with time, that is, with the decreases in gas turbulence are seen in the center region. The opposite tendency is seen for the Lagrangian time scales in the buffer region although the changes are small. [Preview Abstract] |
Monday, November 20, 2017 5:36PM - 5:49PM |
L36.00008: Investigation of particle lift off in a turbulent boundary layer Diogo Barros, Yi Hui Tee, Nicholas Morse, Ben Hiltbrand, Ellen Longmire Entrainment and suspension of particles within turbulent flows occur widely in environmental and industrial processes. Three-dimensional particle tracking experiments are thus conducted in a water channel to understand the interaction of finite-size particles with a turbulent boundary layer. A neutrally buoyant sphere made of wax and iron oxide is first held in place on the bounding surface by a magnet before being released and tracked. The sphere is marked with dots to monitor rotation as well as translation. By setting up two pairs of cameras in a stereoscopic configuration, the trajectories of the sphere are reconstructed and tracked over a distance of 4 to 6$\delta $. Sphere diameters ranging from 40 to 130 wall units, initial particle Reynolds numbers of 600 to 2000 and friction Reynolds numbers of 500 to 1800 are considered. For this parameter set, the particle typically lifts off from the wall after release before falling back toward the wall. Aspects of both particle rotation and translation will be discussed. [Preview Abstract] |
Monday, November 20, 2017 5:49PM - 6:02PM |
L36.00009: Numerical Study of Charged Inertial Particles in Turbulence using a Coupled Fluid-P$^3$M Approach Yuan Yao, Jesse Capecelatro Non-trivial interactions between charged particles and turbulence play an important role in many engineering and environmental flows, including clouds, fluidized bed reactors, charged hydrocarbon sprays and dusty plasmas. Due to the long-range nature of electrostatic forces, Coulomb interactions in systems with many particles must be handled carefully to avoid $O(N^{2})$ computations. The particle-mesh (PM) method is typically employed in Eulerian-Lagrangian (EL) simulations as it avoids computing direct pairwise sums, but it fails to capture short-range interactions that are anticipated to be important when particles cluster. In this presentation, the particle-particle-particle-mesh (P$^3$M) method that scales with $O(N\log{N})$ is implemented within a EL framework to simulate charged particles accurately in a tractable manner. The EL-P$^3$M method is used to assess the competition between drag and Coulomb forces for a range of Stokes numbers and charges. Simulations of like- and oppositely-charged particles suspended in a two-dimensional Taylor-Green vortex and three-dimensional homogeneous isotropic turbulence are reported. One-point and two-point statistics obtained using PM and P$^3$M are compared to assess the effect of added accuracy on collision rate and clustering. [Preview Abstract] |
Monday, November 20, 2017 6:02PM - 6:15PM |
L36.00010: Contact Electrification of Suspended Particles in a Turbulent Fluid Xing Jin, Jeffrey Marshall Contact electrification is a commonly observed phenomenon in which particles exchange charge during collisions in turbulent particulate flows. In various process industries, such as coal mining, sawmills, grain mills and storage facilities, contact electrification is known to lead to dangerous explosions of dust clouds. In natural particulate flow processes, such as sandstorms, volcanic eruptions, planetary rings, and ice transport within thunderstorms, contact electrification leads to development of electrical gradients, often resulting in lightning. In the current work, a probabilistic version of a well-known phenomenological model for contact electrification is used to examine the effect of fluid turbulence on charge development for suspended particles as a function of the particle Stokes number. The distribution of particle collisions and particle charge appear to approach asymptotic states for high values of the Kolmogorov-scale Stokes numbers, exhibiting approximately normal distributions. The influences on particle contact electrification of differences in initial charge carrier density and in particle size are examined. [Preview Abstract] |
Monday, November 20, 2017 6:15PM - 6:28PM |
L36.00011: Marine floc strength and breakup response in turbulent flow Matthew Rau, Steven Ackleson, Geoffrey Smith The effect of turbulence on marine floc formation and breakup is studied experimentally using a recirculating breakup facility. Flocs of bentonite clay particles are grown in a large, stirred aggregation tank of salt water (salinity of 10 ppt) before being subjected to fully-developed pipe flow. Pipe flow conditions range from laminar to turbulent with dissipation rates up to 2.1 m$^{\mathrm{2}}$/s$^{\mathrm{3}}$. Particle size distributions are measured through in-situ sampling of the small-angle forward volume scattering function and through microscopic imaging. Floc size is compared before and after exposure to turbulence and found to be a strong function of the dissipation rate of turbulent kinetic energy. Hydrodynamic conditions within the aggregation tank have a large influence on overall floc strength; flocs formed with stirred aggregation resist breakup compared to flocs formed without stirring. Floc shape and structure statistics are quantified through image analysis and the results are discussed in relation to the measured floc breakup response. Finally, the relevance of these findings to quantifying and predicting marine floc dynamics and the eventual fate of particles in the ocean is presented. [Preview Abstract] |
Monday, November 20, 2017 6:28PM - 6:41PM |
L36.00012: Stochastic four-way coupling of gas-solid flows for Large Eddy Simulations Thomas Curran, Fabian Denner, Berend van Wachem The interaction of solid particles with turbulence has for long been a topic of interest for predicting the behavior of industrially relevant flows. For the turbulent fluid phase, Large Eddy Simulation (LES) methods are widely used for their low computational cost, leaving only the sub-grid scales (SGS) of turbulence to be modelled. Although LES has seen great success in predicting the behavior of turbulent single-phase flows, the development of LES for turbulent gas-solid flows is still in its infancy. This contribution aims at constructing a model to describe the four-way coupling of particles in an LES framework, by considering the role particles play in the transport of turbulent kinetic energy across the scales. Firstly, a stochastic model reconstructing the sub-grid velocities for the particle tracking is presented. Secondly, to solve particle-particle interaction, most models involve a deterministic treatment of the collisions. We finally introduce a stochastic model for estimating the collision probability. All results are validated against fully resolved DNS-DPS simulations. The final goal of this contribution is to propose a global stochastic method adapted to two-phase LES simulation where the number of particles considered can be significantly increased. [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