DESIGN OF AN EFFECTIVE WATER-ALTERNATING-GAS(WAG) INJECTION PROCESS USING ARTIFICIAL EXPERT SYSTEMS

Open Access
Author:
Ma, Jaehyun
Graduate Program:
Petroleum and Natural Gas Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
April 20, 2010
Committee Members:
  • Turgay Ertekin, Thesis Advisor
Keywords:
  • artificial neural network
  • WAG
  • inverse network
Abstract:
Numerical reservoir simulators are commonly used to simulate the water-alternating-gas process for hydrocarbon reservoir. However, a reservoir simulation study can only provide information about the expected reservoir performance and also be time-consuming work. This research presents a screening tool to suggest a set of design parameters that can optimize the WAG process. A commercial reservoir simulator and the neural network toolbox of publicly available software are used in the study. A number of different five spot scenarios were simulated using a commercial program. These scenarios are associated with different reservoir conditions that include initial water saturation, formation thickness, porosity and permeability. They are also involved various design parameters, that is, well spacing, water-gas ratio, alternating frequency, alternating slug size and bottom hole pressure. The reservoir performance is expressed with recovery efficiency and abandonment time. These simulation results and functional links are used to build an artificial neural network of water-alternating-gas process. In this study, an artificial neural network is implemented to construct neuro-simulation tools for screening and designing water-alternating-gas process. These tools generated in this study are effectively able to recognize the connection between the reservoir characteristics and hydrocarbon production performance of the WAG process in order to forecast proper operating conditions. They can also serve to provide a relatively narrow range of possible scenarios and reduce the time for conventional reservoir simulations.