Abstract
dc:description.abstractThe work presented considers direct numerical simulation (DNS) based studies of bubbly flows in turbulent channel and pipe flows. A variety of bubble sizes are simulated by means of Lagrangian tracking (LPT) and the volume of fluid (VOF) approaches to gain an in-depth knowledge of the physical phenomena involved in, and provide a predictive capability for, turbulent bubbly flows. DNS coupled with LPT is used to investigate microbubble dynamics in four-way coupled channel and pipe flows. Microbubbles, assumed to be non-deformable and spherical, in channel flows show that one- and two-way coupled predictions demonstrate different trends, with microbubbles pushed by the lift force towards the channel wall in upflow and towards its centre in downflow. Analysis of bubble collision behaviour highlights that binary collisions most frequently occur at very small approach angles and with low relative approach velocities. This trend is confirmed in pipe flows, with bubble coalescence in both geometries predicted using film drainage and energy-based models. DNS of turbulent channel flow with large deformable bubbles is studied using the VOF method. The motion of single bubbles is considered, with 8 mm bubbles travelling in roughly rectilinear paths in upflow owing to their higher deformability than smaller bubbles which move towards the walls. A method for estimating the drag coefficient is proposed, with good agreement found with experimental data. Time-averaged liquid turbulence statistics are evaluated to quantify bubble-induced turbulence, and bubble clustering is analysed by simulating bubble swarms with a horizontal alignment of colliding bubbles found. The work reported contributes new understanding of the phenomena involved in bubbly flows down to the smallest scales, with the VOF method, in particular, allowing highly accurate results to be generated that improve understanding. The DNS-LPT technique, for the foreseeable future, will remain the main predictive tool for modelling practically relevant bubbly flows.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Leeds
- Year dc:date.issued
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Zhai, Jian
- Advisor dc:contributor.advisor
-
- Fairweather, Michael
- ORCID person:identifier.orcid
Rights
- Language dc:language.dcterms_iso639_2
- eng
Identifiers
dc:identifier.*- Identifier
- uk.bl.ethos.826656