Event Type:

Applied Mathematics and Computation Seminar

Date/Time:

Friday, April 19, 2013 - 05:00

Location:

GLK 104

Event Link:

Guest Speaker:

Justin Finn

Institution:

OSU Mechanical Engineering

Abstract:

Despite immediate relevance to both natural and engineered systems such as

groundwater networks and chemical reactors, flows through packed beds of spheres

are not generally well understood. This is particularly true at moderate to high

Reynolds numbers where fluid inertia results in complex, three dimensional flow

features including jets and vortices. Such porescale features can have implications

for macroscale properties of engineering interest including pressure drop, dispersion,

and reaction rates.

In this work, direct numerical simulation (DNS) was used to investigate the porescale

structure of flow through synthetic sphere packings at moderate Reynolds numbers,

between 10 an 600. To first choose and validate appropriate computational models

for this problem, the relative performance of two numerical approaches involving body

conforming and non-conforming grids was examined in detail. Next, data from

simulations in the steady and unsteady inertial flow regimes was used to analyze the

characteristics of porescale vortical structures. Even at similar Reynolds numbers,

the structure and behavior of vortices observed in arranged and random packings

are remarkably different. Finally, a numerical tool was developed to study packed

bed flows using the recently developed theory of Lagrangian Coherent Structures

(LCS). LCS are invariant barriers to transport and define dynamically distinct regions

of time dependent flows. The computations were embedded directly in the DNS

solver, allowing LCS theory to be applied to complex three dimensional problems,

including packed beds, for the first time.