Degree Type

Dissertation

Date of Award

2013

Degree Name

Doctor of Philosophy

Department

Mechanical Engineering

First Advisor

Shankar Subramaniam

Abstract

Gas-solid flows are commonly encountered in Nature and in several industrial applications. Emerging carbon-neutral or carbon negative technologies such as chemical looping combustion and CO2 capture are examples of gas-solid flows in power generation industry. Computational fluid dynamics (CFD) simulations are increasingly being seen as a cost-effective tool in the design of technological applications in power generation industry. Device-scale CFD calculations that involve gas-solid flow are based on statistical descriptions that require closure models for the exchange of mass, momentum, energy and heat transfer between the dispersed solid phase and the gas phase. The predictive capability of multiphase flow CFD simulations strongly depends on the accuracy of the models used for the interphase exchange terms. Particle-resolved direct numerical simulation (PR-DNS) is a first-principles approach to develop accurate models for interphase momentum, energy and heat transfer in gas-solid flow. The primary objective of this work is the development of accurate models for the interphase exchange of momentum, kinetic energy and heat transfer in polydisperse gas-solid flows using PR-DNS.

A novel computational tool named Particle-resolved Uncontaminated-fluid Reconcilable Immersed Boundary Method (PUReIBM) has been developed as a part of this work to perform PR-DNS of flow past fixed and freely moving spherical particles. We designed the appropriate numerical experiment that can be used to develop closure models for interphase momentum transfer and formally established the connection between PR-DNS and statistical theory of multiphase flow for which the models are intended. Using PUReIBM we developed an improved drag correlation to model interphase momentum transfer in gas-solid flow. The solution fields obtained from PUReIBM PR-DNS have been used to quantify the velocity fluctuations in the gas-phase and a simple eddy viscosity model for the gas-phase pseudo-turbulent kinetic energy has been developed. A novel PR-DNS methodology to study heat transfer in gas-solid flow has been developed. These results provide insight into the role of fluid heating in gas-solid flow and motivate the development of better models for gas-solid flow heat transfer. From PR-DNS of freely evolving gas-solid suspensions we developed a stochastic model for particle acceleration that accounts for the particle velocity distribution. In addition to model development, the implementation of a parallel algorithm that enables PR-DNS of gas-solid flow on petascale supercomputers is also discussed.

DOI

https://doi.org/10.31274/etd-180810-3347

Copyright Owner

Sudheer Tenneti

Language

en

File Format

application/pdf

File Size

272 pages

Share

COinS