Permeability of fracture networks generated through geomechanical fracture-growth simulations
File(s)
Author(s)
Thomas, Robin Nicholas
Type
Thesis
Abstract
Fracture networks in subsurface rocks significantly impact the mechanical and hydraulic properties of the Earth. Realistic models of fracture networks underpin a variety of important geoscience fields. The physics of how fractures propagate and interact control how networks develop, but such processes are not routinely incorporated into discrete fracture network (DFN) models. This thesis presents a novel approach to model deformation of isotropic, homogeneous, and linear elastic brittle rocks, leading to the development of dense three-dimensional fracture networks. These networks provide realistic representations of subsurface networks that honour the physical process of concurrent fracture growth and interaction. This thesis proposes a number of novel developments that advance the field of discrete fracture modelling, with the aim of growing dense networks in a realistic manner. It presents a number of new validations for fracture growth and interaction, providing benchmarks for the accuracy required to generate realistic networks. Towards the goal of generating realistic fracture networks, several topics are studied numerically. A method for examining fracture interaction in three-dimensions is introduced, which quantifies the change in stress a fracture undergoes due to a neighborhood of nearby fractures. Fracture interaction is examined spatially in novel fracture interaction maps. Realistic fracture growth is demonstrated in both tension and compression. The effect of fracture interaction on mutual fracture growth is demonstrated on fracture arrays. The hydraulic properties of a spent UO\textsubscript{2} fuel pellet are quantified using an effective permeability computation method. The thesis concludes by presenting six geomechanical discrete fracture networks (GDFNs). The primary influences on the geometric and hydraulic properties of these GDFNs are the stresses causing growth and fracture interaction, rather than the random process of initial fracture insertion. GDFNs provide a new way of understanding subsurface processes, and improve our capacity to model hydro-mechanical processes in fractured rock masses.
Version
Open Access
Date Issued
2019-03
Date Awarded
2019-11
Copyright Statement
Creative Commons Attribution NonCommercial Licence
Advisor
Paluszny, Adriana
Zimmerman, Robert W
Sponsor
Natural Environment Research Council (Great Britain)
European Commission
Grant Number
NE/L000660/1
309067
GA/12F/101
Publisher Department
Earth Science & Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)