Justin Finn

A Variable-Density Fictitious Domain Method for Particulate Flows with Broad Range of Particle-Fluid Density Ratios

A numerical scheme for fully resolved simulation of uid-particle systems with freely moving rigid particles is developed. The approach is based on a ctitious domain method wherein the entire uid-particle domain is

Modeling and simulation of multiple bubble entrainment and interactions with two dimensional vortical flows

Simulations of bubble entrainment and interactions with two dimensional vortical flows are preformed using a discrete element model. In this Eulerian-Lagrangian approach, solution to the carrier phase is obtained using direct numerical simulation whereas motion of subgrid bubbles is modeled using Lagrangian tracking. The volumetric displacement of the fluid by the finite size of the bubbles is modeled along with interphase momentum-exchange for a realistic coupling of the bubbles to the carrier phase. In order to assess the importance of this volumetric coupling effect, even at low overall volume loading, simulations of a small number of microbubbles entrained in a traveling vortex tube are studied in detail. The test case resembles the experiments conducted by Sridhar and Katz [JFM, 1999] on bubble entrainment in vortex rings. It is shown that under some conditions, the entrainment of eight small bubbles, 1100 μm or less in diameter, result in significant levels of vortex distortion when modeled using th...

Integrated computation of finite-time Lyapunov exponent fields during direct numerical simulation of unsteady flows

The computation of Lagrangian coherent structures typically involves post-processing of experimentally or numerically obtained fluid velocity fields to obtain the largest finite-time Lyapunov exponent (FTLE) field. However, this procedure can be tedious for large-scale complex flows of general interest. In this work, an alternative approach involving computation of the FTLE on-the-fly during direct numerical simulation of the full three dimensional Navier-Stokes equations is developed. The implementation relies on Lagrangian particle tracking to compose forward time flow maps, and an Eulerian treatment of the backward time flow map [S. Leung, J. Comput. Phys. 230, 3500-3524 (2011)] coupled with a semi-Lagrangian advection scheme. The flow maps are accurately constructed from a sequence of smaller sub-steps stored on disk [S. Brunton and C. Rowley, Chaos 20, 017503 (2010)], resulting in low CPU and memory requirements to compute evolving FTLE fields. Several examples are presented to demonstrate the capability and parallel scalability of the approach for a variety of two and three dimensional flows.

MODELING AND SIMULATION OF MULTIPLE BUBBLE ENTRAINMENT AND INTERACTIONS WITH A TRAVELING VORTEX RING

Bubble interactions with vortical structures are important to better understand the mechanisms of bubble induced boundary layer drag reduction and chemical mixing. Traditionally, many studies of disperse bubble or particle-laden flows have utilized an Euler-Lagrange two-way coupling approach, wherein the dispersed phase is assumed subgrid and its dynamics is modeled. In this work, results on full three-dimensional simulation of traveling vortex ring together with a few microbubbles are presented utilizing a volumetric coupling approach, wherein the displaced mass due to the presence of the bubbles is accounted for by using mixture theory based conservation laws in an Euler-Lagrange formulation. It is shown that the volumetric coupling approach is necessary to reproduce the experimental observations of Sridhar & Katz, JFM (1999). Experimental work by S&K on bubble entrainment into a traveling vortex ring has shown that the settling location of the bubble relative to the vortex core can be well predicted based on the ratio of the buoyancy force to the hydrodynamic pressure gradient. Additionally, the experimental results find that even at low volume fractions, bubble injection can significantly affect the structure of the vortex core. The two-way coupling model, wherein the fluid displacement due to bubble motion is neglected, of bubble-laden flows is unable to capture these effects on the vortical structure.Copyright

Characteristics of Porescale Vortical Structures in Random and Arranged Packed Beds of Spheres

The characteristics of pore scale vortical structures observed in moderate Reynolds number flow through mono-disperse packed beds of spheres are examined. Our results come from direct numerical simulations of flow through (i) a periodic, simple cubic arrangement of 54 spheres, (ii) a wall bounded, close packed arrangement of 216 spheres, and (iii) a realistic randomly packed tube containing 326 spheres with a tube diameter to sphere diameter ratio of 5.96. Pore Reynolds numbers in the steady inertial (10 ≲ Re ≲ 200) and unsteady inertial (Re ≈ 600) regimes are considered. Even at similar Reynolds numbers, the vortical structures observed in flows through these three packings are remarkably different. The interior of the arranged packings are dominated by multi-lobed vortex ring structures which align with the principal axes of the packing. The random packing and the near wall region of the close packed arrangement are dominated by helical vortices, elongated in the mean flow direction. In the simple cubic packing, unsteady flow is marked by periodic vortex shedding which occurs at a single frequency. Conversely, at a similar Reynolds number, the vortical structures in unsteady flow through the random packing oscillate with many characteristic frequencies.Copyright

Relative Performance of Body-Fitted and Fictitious Domain Simulations of Flow Through Porous Media

The relative performance of (i) a body-fitted unstructured grid Navier-Stokes solver [Moin and Apte, AIAA J. 2006], and (ii) a fictitious domain based finite-volume approach [Apte et al. JCP 2009] is examined for simulating flow through packed beds of spheres at moderate flow rates, 50 ≲ Re ≲ 1300. The latter employs non-body conforming Cartesian grids and enforces the no-slip conditions on the pore boundaries implicitly through a rigidity constraint force. At these flow rates, fluid inertia can result in complex steady and unsteady pore scale flow features that influence macro-scale properties. We examine the requirements on both methods to properly capture these features in both simple and complex arrangements of spheres. First, two prototypical test cases of flow through packed beds are studied thoroughly at a range of Reynolds numbers in the inertial flow regime. Next flow through a random packing of 51 spheres at Re = 1322 is simulated using both methods. The suitability of both approaches to the complex configurations observed in large randomly packed beds is discussed.Copyright

EXPERIMENTAL VERSUS COMPUTATIONAL METHODS IN THE STUDY OF FLOW IN POROUS MEDIA

Both experimental and computational methods applied to the study of porous media flows are challenging due to the complex multi-phase geometry and ability to resolve scales over a reasonably large domain. This study compares experimentally obtained results based on refractive index matching of detailed velocity field vectors with those obtained using DNS to evaluate both methods for consistency. Data were obtained in a randomly packed bed using uniformly sized spherical particles. Experimental challenges including refractive index matching errors, magnification uncertainties, and the identification of the proper geometry as well as, the arduousness, of matching the geometry, grid resolution particularly near solid contact points, and proper boundary conditions DNS are presented. Detailed comparison of the numerical simulation with PIV measurements are presented by attention paid to the statistical distribution of velocities, and their deviation from DNS estimations from the measured values. There is reasonable matching the velocity fields except for some regions of constricted flow. The axial velocity results are within 12 percent and the normal velocity within 9%. Streamline details show that both methods agree well. It is found that inlet conditions play a significant role in being able to match results. NOMENCLATURE

Numerical Simulation of Sand Ripple Evolution in Oscillatory Boundary Layers

We perform simulations of sand ripple evolution in an oscillatory boundary layer flow typical of the ripple regime. The simulation framework is a parallel implementation of a three dimensional, variable density, incompressible flow solver, which solves the ensemble averaged Navier-Stokes equations on a fixed, structured grid. The sediment phase is evolved by computing hydrodynamic and inter-particle forces acting on each Lagrangian particle. Particle-particle collisions are treated with a soft sphere model incorporating both normal and tangential collision forces. Realistic and consistent coupling of the sediment to the Eulerian fluid phase is achieved through a typical inter-phase drag force term as well as the effects of volume displacement by the sediment. The Euler-Lagrange computational approach is developed in three-dimensions and its accuracy is verified using two test cases with analytic or empirically known solutions. It is then applied to simulate ripple evolution in oscillatory boundary layers and results are compared with Nielsens ripple predictor model as well as mixture-theory based Eulerian computations.Copyright