The Development of an Artificial Neural Network as a Pressure Transient Analysis Tool with Application to Multi-lateral Wells in Tight-gas Dual-porosity Reservoirs
Open Access
Author:
Cox, Jacob Steven
Graduate Program:
Petroleum and Natural Gas Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
May 15, 2014
Committee Members:
Turgay Ertekin, Thesis Advisor/Co-Advisor
Keywords:
ANN Artificial Neural Network Tight-Gas Dual-Porosity Multi-Lateral Pressure Transient
Abstract:
The goal of this study was to create a tool with the use of an artificial neural network (ANN) that could quickly predict the inverse solution to pressure transient (PT) data created for multiple multi-lateral wells completed within a tight-gas dual-porosity reservoir. This inverse tool would be able to predict the user’s reservoir parameters nearly instantaneously with known inputs of PT data and wellbore design. This tool will take ideas from current well test analysis to aid in the training of the neural network. However, once the network has been trained, it will be able to predict multiple key reservoir properties and the time consuming process of conventional well test analysis will no longer be an issue.
