Semester

Spring

Date of Graduation

2011

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Petroleum and Natural Gas Engineering

Committee Chair

Shahab D Mohaghegh

Abstract

The work presented in this thesis is an intelligent reservoir modeling and analysis of Lower Huron Shale in eastern Kentucky. Methodology used for this analysis is a recently developed Top Down Intelligent Reservoir Modeling, that couples artificial intelligence and data mining techniques with conventional reservoir engineering methods. A total of 77 wells completed in Lower Huron Shale in eastern Kentucky were used in this study. Well production data was obtained from the company operating the wells, while completion reports and well logs were downloaded from publicly available database at Kentucky Geological Surveys website. The downloaded well logs were digitized, and detailed geological interpretation of the studied area was performed. More information about the reservoir was acquired through decline and type curve analyses and single well history matching. Single well history matching was performed with publicly available shale and tight gas reservoir simulator. All of the acquired data was used in the development of spatiotemporal dataset that was further analyzed with state of the art in artificial intelligence and data mining (fuzzy pattern recognition, artificial neural networks). Finally, reservoir was divided into zones of different relative reservoir quality and full field artificial intelligence empowered predictive model of the reservoir was developed and verified.

Share

COinS