ARTIFICIAL NEURAL NETWORK BASED DESIGN PROTOCOL FOR WAG IMPLEMENTATION IN CO2 INJECTION USING FISHBONE WELLS IN LOW PERMEABILITY Oil RESERVOIRS

Open Access
Author:
Enab, Khaled Ahmed
Graduate Program:
Energy and Mineral Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
December 07, 2017
Committee Members:
  • Turgay Ertekin, Dissertation Advisor
  • Turgay Ertekin, Committee Chair
  • Sanjay Srinivasan, Committee Member
  • Gregory R King, Committee Member
  • Sridhar Anandakrishnan, Outside Member
Keywords:
  • Enhanced Oil Recovery
  • WAG
  • CO2 Injection
  • Unconventional Oil Reservoirs
  • Tight Oil Reservoirs
  • Artificial Neural Network
  • Machine Learning
Abstract:
Increasing oil recovery and decreasing production costs in unconventional reservoirs is in great demand due to current low oil prices. The oil price drop affected unconventional oil production to a greater degree than conventional oil production because of its low production rates and high production costs. The objective of increasing oil recovery and decreasing production costs of the unconventional oil reservoirs leads to an introduction of new technology and techniques. Many studies have been conducted to identify new techniques with high capabilities of increasing oil production from unconventional resources. Unconventional oil reservoirs are the reservoirs that cannot feasibly produce oil using conventional production techniques. These reserves include tight oil, oil shale, and bitumen. The first objective of this work is to introduce a new production technique that is capable of increasing oil recovery, decreasing production costs for unconventional oil reservoirs, and decreases the greenhouse gas emissions. The second objective is to build an Artificial Neural Network (ANN) based toolbox to evaluate and optimize the implementation of the proposed technique. The proposed ANN toolbox provides the necessary knowledge to understand the performance of the CO2 injection technique in low permeability reservoirs when utilizing fishbone well designs. A better understanding of how the implementation of the fishbone well design and CO2-WAG injection affects the production from low permeability reservoirs, will help to increase oil recovery, decrease production costs, and decrease environmental impacts. A better understanding of how the implementation of the fishbone well design and CO2-WAG injection affects the production from low permeability reservoirs will help to increase oil recovery, decrease production costs, and decrease environmental impacts. The proposed production technique combines two successfully applied techniques in the unconventional reservoir production. The first technique is the Water Alternate Gas (WAG) injection technique implementation with carbon dioxide. The second technique is the fishbone multilateral well designs. The proposed techniques have successfully proven their capability of increasing oil production from unconventional reservoirs, but there are no previous studies that combined both techniques in the same project. Water alternate gas is a known term that describes an enhanced oil recovery process whereby water injection and gas injection are carried out alternately for periods of time to provide better sweep efficiency and reduce gas channeling from injector to producer. Carbon dioxide (CO2) is the most commonly used gas for this process because it improves hydrocarbon contact time and sweep efficiency while decreases the greenhouse gas emissions. The fishbone well design is a multi-lateral well technique which has been successfully used to improve production from low permeability reservoirs. The fishbone well design is a series of multilateral well segments that trunk off a main horizontal well. The appearance of the fishbone well closely resembling the ribs of a fish skeleton deviated from the main backbone. Fishbone well design increases the production by increasing the contact area with the reservoir. The utilization of the multi-lateral well technique increases the cumulative fluid recovery, decreases the environmental footprint and decreases the drilling and completion costs. An Artificial Neuron Network (ANN) is a computational model based on the structure and functions of biological neural networks. Information that flows through the network affects the structure of the ANN due to the adaptions it makes to provide the desired output. ANNs are considered nonlinear statistical data modeling tools where the complex relationships between inputs and outputs are modeled. ANN toolbox has been developed to evaluate and optimize the implementation of CO2-WAG injection in low permeability reservoirs using fishbone well design. The developed toolbox provides the needed information to develop a successful production plan for the reservoir under consideration. The designed toolbox has four primary functions. The first function is to predict the fluid flow rate profiles for the reservoir under consideration, with fishbone well and WAG injection implementation. The second function is to provide a reliable WAG injection design for a project with known reservoir properties while also creating a functional fishbone well after the field goes through a period of primary production. This will achieve the desired level of recovery for the project. The third function is to provide a reliable project design for the reservoir under consideration. The project design consists of a fishbone well design and a WAG injection design to achieve a desired level of recovery. The fourth function is to understand the reservoir properties using the production data. The toolbox can compare thousands of WAG designs and multilateral fishbone well design combinations in a more rapid manner when compared to commercial simulators.