Date of Graduation

1998

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Petroleum and Natural Gas Engineering

Committee Chair

Shahab Mohaghegh

Abstract

The purpose of this study is to develop a Front-End and an Optimization Routine that will help petroleum engineers to design optimum treatment schedules for hydraulic fracturing. This tool reduces the time engineers spend designing optimum treatment schedules by proposing a solution based on virtual intelligence. Neural networks and genetic algorithms are integrated in a software application to achieve the objective of this study. A methodology to reduce the total number of layers to four characteristic layers was developed. Using this methodology, two types of neural networks were trained. A new selection method was developed for the genetic algorithm process. The methodologies coupled with neural network and genetic algorithm were integrated in a user-friendly application compiled in a Visual Basic programming language. The results show the usefulness of the software in designing treatments that create desired propped length. The outputs were found to be an excellent fit when compared with the original treatments.

Download

Share

COinS