Thesis

Development and validation of a novel hybrid CFD slurry model

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2019
Thesis identifier
  • T15172
Person Identifier (Local)
  • 201479809
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • Erosion from slurry flows is a large problem in the mining and oil and gas industries, hence the interest from Weir group. Centrifugal slurry pumps manufactured by the Weir group are considered to be some of the world's best, however the harsh environments they operate in can erode hardened steel parts in a matter of weeks. One way to mitigate the damage caused to the parts is better design. This is achieved by simulating the flow with computational fluid dynamics (CFD), using a two phase model: one phase for the fluid, one phase for the particles.;The two phase models can, broadly speaking, be separated into two categories: Euler-Euler (EE), and Euler-Lagrange (EL). EE models the fluid and particles as two continua, is fast to run, but lacks particle impact data which is required for erosion modelling. EL models a fluid phase and a particulate phase, is more computationally expensive than the EE model, but has the benefit of having particle impacts at the wall.;A hybrid slurry model was developed combining both EE and EL models, taking advantage of each model's strengths. The hybrid model proved that a combined EE/EL solver was capable of modelling slurry flows in less time than a pure EL model, but with more particle impact data at the wall than a EE model. Issues with combining two computational models in one domain were identified and overcome, buy including a transition baffle, where the solver changed from EE to EL.;Experiments were carried out to validate the computational simulation results on a submerged jet impingement test. Particle image velocimetry was carried out to obtain data from the experiments, and images were correlated to output particle velocity data.
Advisor / supervisor
  • Stickland, Matthew
  • Dempster, Bill
Resource Type
DOI
Date Created
  • 2019
Former identifier
  • 9912691391702996

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T15172 2021-07-02 Public Download