ERA

Download the full-sized PDF of Analyzing Scaling Characteristics of Transport Properties  Using Particle-Tracking Based TechniquesDownload the full-sized PDF

Analytics

Share

Permanent link (DOI): https://doi.org/10.7939/R3X63BK44

Download

Export to: EndNote  |  Zotero  |  Mendeley

Communities

This file is in the following communities:

Graduate Studies and Research, Faculty of

Collections

This file is in the following collections:

Theses and Dissertations

Analyzing Scaling Characteristics of Transport Properties Using Particle-Tracking Based Techniques Open Access

Descriptions

Other title
Subject/Keyword
Non-Fickian transport
Scaling characteristics
Geostatistics
Sub-scale variability
Statistical scale-up
Scale dependency
Particle tracking
Multi-phase particle tracking
Multi-scale heterogeneity
Dispersion
Type of item
Thesis
Degree grantor
University of Alberta
Author or creator
Vishal, Vikrant
Supervisor and department
Juliana Y. Leung (Petroleum Engineering)
Examining committee member and department
Japan Trivedi (Petroleum Engineering)
Clayton Deutsch (Mining Engineering)
Edwin Cey (External, University of Calgary)
Hassan Dehghanpour (Petroleum Engineering)
Department
Department of Civil and Environmental Engineering
Specialization
Petroleum Engineering
Date accepted
2017-09-27T13:27:48Z
Graduation date
2017-11:Fall 2017
Degree
Doctor of Philosophy
Degree level
Doctoral
Abstract
Appropriate scale-up provides a critical link between fine-scale heterogeneity descriptions and coarse-scale models used for transport modeling, which is essential for planning and management of subsurface reservoirs. A significant challenge in subsurface flow and transport modeling is to develop scale-appropriate parameters to represent physical heterogeneities that impact solute migration and flow response. Another challenge is to construct reservoir models that would capture the uncertainties stemming from incomplete data (often gathered over different scales) and loss of information or smoothing due to averaging. Fine-scale models contain detailed descriptions of reservoir properties, but these models can be too computationally demanding and are not practically feasible for routine reservoir simulation. Coarse-scale models often offer a viable alternative that could decrease computational demand substantially. However, the increased grid-block size in the coarse scale model leads to an increase in numerical (or artificial) dispersion, which stems from the truncation error from most numerical discretization schemes and is directly proportional to grid-block size. The main issue with numerical dispersion when examining scale-up characteristics is that it tends to overwhelm the physical (or actual) dispersion. Alternative transport modeling schemes, such as the Lagrangian (particle-tracking) methods, are widely adopted in simulating solute transport in porous media. Its primary advantage over typical numerical discretization methods (e.g., finite volume) is the absence of numerical dispersion and potential computational efficiency. More importantly, certain particle-tracking methods are capable of modeling this type of anomalous behavior of transport. In this research, a new particle-tracking method is developed for simulating probabilistic (or random) transition time steps and multi-phase immiscible flow. This is further integrated in a novel hierarchical framework for scale-up of reservoir and transport model parameters including porosity, dispersivity, and multi-phase flow functions (e.g., relative permeability and capillary pressure). A key feature of the developed particle-tracking formulation is the employment of kernel estimator for computing concentration and saturation distribution, which has greatly improved the overall computational efficiency by reducing the number of particles needed to achieve a consistent distribution. The developed particle-tracking method for both probabilistic transition time steps and multi-phase immiscible flow is validated against the analytical solution and is demonstrated to alleviate numerical dispersion when compared against common numerical discretization (e.g. finite difference) methods. Predictions obtained from the coarse-scale models constructed according to the developed workflow are shown to be more consistent with the fine-scale model.
Language
English
DOI
doi:10.7939/R3X63BK44
Rights
This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for the purpose of private, scholarly or scientific research. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.
Citation for previous publication

File Details

Date Uploaded
Date Modified
2017-09-27T19:27:49.449+00:00
Audit Status
Audits have not yet been run on this file.
Characterization
File format: pdf (PDF/A)
Mime type: application/pdf
File size: 9570328
Last modified: 2017:11:08 17:58:57-07:00
Filename: Vishal_Vikrant_201709_PhD.pdf
Original checksum: 15e39919ee3957f43db08d6eda2185d6
Activity of users you follow
User Activity Date