Download the full-sized PDF of Real-Time Feedback Control of the SAGD Process using Model Predictive Control to Improve Recovery: A Simulation StudyDownload the full-sized PDF



Permanent link (DOI):


Export to: EndNote  |  Zotero  |  Mendeley


This file is in the following communities:

Graduate Studies and Research, Faculty of


This file is in the following collections:

Theses and Dissertations

Real-Time Feedback Control of the SAGD Process using Model Predictive Control to Improve Recovery: A Simulation Study Open Access


Other title
Production Optimization
Smart Fields
Intelligent Wells
Steam Assisted Gravity Drainage
Smart Wells
Model Predictive Control
Advanced Process Control
System Identification
Subcool Control
Feedback Control
Real-Time Optimization
Type of item
Degree grantor
University of Alberta
Author or creator
Vembadi, Shiv S.
Supervisor and department
Trivedi, Japan (Department of Civil and Environmental Engineering)
Prasad, Vinay (Department of Chemical and Materials Engineering)
Examining committee member and department
Shah, Sirish (Department of Chemical and Materials Engineering)
Prasad, Vinay (Department of Chemical and Materials Engineering)
Li, Huazhou (Department of Civil and Environmental Engineering)
Trivedi, Japan (Department of Civil and Environmental Engineering)
Department of Civil and Environmental Engineering
Petroleum Engineering
Date accepted
Graduation date
Master of Science
Degree level
Scope of the Work: For over a decade, the oil industry has been moving to “smart fields”, which deploy wells with remotely operated valves and permanently installed downhole sensors for real-time pressure and temperature measurements. Real-time data from these “intelligent wells” provide key knowledge about the reservoir performance and enable continuous and automatic production optimization for better economics. In this study, we apply intelligent wells with fiber-optic array temperature sensing to Steam Assisted Gravity Drainage (SAGD) for real-time production optimization using Model Predictive Control (MPC), which is a multivariable constrained control strategy. A linear empirical model is first identified using downhole temperature and well rate data. Based on the linear model and real-time temperature and rate data, an MPC controller manipulates the well rates to control the subcool along a well pair in a SAGD reservoir. We use a multilevel control framework, in which the well settings from long-term optimization using a reservoir model provide the “set points” for MPC. Procedure: To evaluate the use of MPC for real-time control of subcool in SAGD, we use three-dimensional heterogeneous reservoir models with a single pair of dual tubing string horizontal wells. A set of porosity and permeability realizations are created. Two realizations are selected to represent two different cases of uncertain reservoir models. Further, another realization is created that is considered as the “synthetic” (virtual) reservoir. For each of the two reservoir models, a proprietary reservoir simulator is used to find the optimum rates and subcool. Then MPC is used to control the subcool along the well pair in synthetic reservoir. Results, Observations and Conclusions: Using the multilevel control framework, NPV improves by 18.23% and 8.81% in the two cases of reservoir models, over a direct application of the optimum rates. Though the results validate the use of MPC for real-time optimization of SAGD, we faced a couple of issues (which have related practical concerns) in the identification of good linear models and subsequently using them in the MPC controller because of steam breakthrough in the dual tubing string well pair. However, we conclude that identification of good linear models will be feasible if ICVs are used in the injector and producer, which allow for more uniform steam distribution in the injector and differential steam trap control in the producer. Work’s Novelty: A few other works present results for the use of proportional-integral-derivative (PID) control for automatic feedback control of the SAGD variables. However, unlike the MPC strategy, the PID-control strategy is a single-input and single-output control strategy. It acts on each controlled variable in isolation by manipulating a single variable instead of optimizing the whole system as the MPC strategy does.
Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
Citation for previous publication

File Details

Date Uploaded
Date Modified
Audit Status
Audits have not yet been run on this file.
File format: pdf (PDF/A)
Mime type: application/pdf
File size: 2448989
Last modified: 2015:10:22 06:10:57-06:00
Filename: Vembadi_Shiv_S_201412_MSc.pdf
Original checksum: 83b5aec632befec7ae6d0970d8f1b00f
Activity of users you follow
User Activity Date