Dissertation/Thesis Abstract

Prediction and Analysis of Geomechanical Properties of the Upper and Middle Bakken Formation Utilizing Artificial Intelligence and Data Mining
by Parapuram, George Kurian, M.S., University of Louisiana at Lafayette, 2017, 110; 10682660
Abstract (Summary)

To efficiently produce oil from unconventional reservoirs, it is imperative to determine and understand the geomechanical properties of the formation. But, due to the high cost of obtaining these properties from geomechanical well logs, businesses are looking for all possible ways to cut cost. The plummeting oil prices have been reflected in company spending and have driven companies to prioritize focusing attention on the rising production costs and venture all possible ways to reduce these costs. The real challenge is how to preserve these profitable gains? There is a need for an alternate and cost- effective way to obtain geomechanical properties of the rocks.

By utilizing Data Analytics, Data Mining, and ANN, patterns are observed between parameters from large amounts of data and, thus, important information regarding the formation can be understood. In this study, a relationship between conventional well logs and geomechanical well logs are established. Properties such as Young’s Modulus, Poisson’s Ratio, Shear Modulus, Bulk Modulus, and Minimum Horizontal Stress are determined from Conventional Logs such as Gamma Ray and Density Log utilizing ANN. Ultimately, data-driven models are developed to predict accurate geomechanical properties for future wells of the Upper and Middle Bakken Formation. Finally, the efficacy of the data-driven models achieved is tested on randomly selected new wells that were not used in the training of the model. The accurate prediction and analysis of these properties help in better reservoir characterization and efficient production from the future wells in the Bakken Formation.

Indexing (document details)
Advisor: Mokhtari, Mehdi
Commitee: Ben Hmida, Jalel, Boukadi, Fathi H., Seibi, Abdennour C.
School: University of Louisiana at Lafayette
Department: Petroleum Engineering
School Location: United States -- Louisiana
Source: MAI 57/05M(E), Masters Abstracts International
Subjects: Petroleum Geology, Petroleum engineering, Artificial intelligence
Keywords: Artificial Neural Networks (ANN), Data analysis, Data mining, Geomechanics, Minimum horizontal stress, Shear wave
Publication Number: 10682660
ISBN: 9780355854305
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy