Semester

Fall

Date of Graduation

2019

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

Committee Co-Chair

Samuel Ameri

Committee Member

Kashy Aminian

Abstract

Data driven reservoir modeling, also known as Top-Down Model (TDM), is an alternative to the traditional numerical reservoir simulation technique. Data driven reservoir modeling is a new technology that uses artificial intelligence and machine learning to build full-field reservoir models using field measurements (data - facts) instead of mathematical formulations that represent our current understanding of the physics of the fluid flow through porous media. TDM combines all field measurements into a comprehensive reservoir model to predict the production from each well in a field with multiple wells.

There are many opinions, speculations and criticism about not using the physics-based approach. Therefore, in this thesis, to confirm the capabilities of TDM, synthetic data generated from a numerical reservoir simulation model will be used for the development of a Data Driven Reservoir Model. That means, the physics of the fluid flow through porous media will be modeled using the generated data from the numerical reservoir simulation model which we know everything about.

In order to accomplish the objectives of this thesis a software application will be used for the development of the Top-Down Model. TDM will be developed (trained, calibrated and validated) and history matched using the data generated by a complex numerical reservoir simulation model in order to confirm the capabilities of the TDM in forecasting existing well behavior. Upon Completion of the TDM, predictions will be made using the developed TDM and are tested against the data that will be generated by the numerical reservoir simulation.

Required Additional Files
Signatures form.pdf (312 kB)
Download

Share

COinS