Machine Learning Applications in Oil Well Petrophysics and Original Oil in Place Estimation: A Systematic Literature Review | Chapter 1 | Current Approaches in Engineering Research and Technology Vol. 8

One type of artificial intelligence (AI) that can be used in any field of study is machine learning. Numerous algorithms that are used to perform different tasks, including classification, estimation, prediction, comparison, approximation, optimization, and selection, are included in it. The reservoir engineer can predict the total amount of crude oil that may be in the reservoir by assessing the original oil in place. When there is limited data available, machine learning is found to execute reserves estimation quickly and accurately. These and other characteristics of machine learning led to a thorough evaluation of the literature covering research done between 2010 and 2021, examining the advantages and disadvantages of the studies as they were presented. To determine the hydrocarbon potential of a reservoir, the oil industry collects several kinds of data from both subsurface and surface sources. Large-scale data collection, analysis, and output prediction are all known to be possible using sensors. After a set of inclusion and exclusion criteria were applied to the 3127 study-related papers that were gathered from 4 databases, 104 journal articles that satisfied the requirements were used for the review. According to the study's findings, 2019 had the most publications (20 out of 104) on the subject under review among the years under evaluation. Additionally, 39% of authors noted that the algorithm was not performing well, and 61% of authors reported insufficient data. It was also discovered that the sector employed machine learning more for forecasting and prediction than for other purposes, with artificial neural networks (ANNs) being the most used AI technology. The work provides researchers with further insights into machine learning for petrophysics analysis and original oil in place estimation. This clever new technology simplifies and streamlines the process of evaluating data.

 

Author (s) Details

Ekemini Anietie Johnson
Department of Computer Science, University of Uyo, Nigeria and TETFund Centre of Excellence in Computational Intelligence Research, University of Uyo, Nigeria.

 

Okure Udo Obot
Department of Computer Science, University of Uyo, Nigeria and TETFund Centre of Excellence in Computational Intelligence Research, University of Uyo, Nigeria.

 

Kingsley Attai
Department of Mathematics and Computer Science, Ritman University, Nigeria.

 

Julius Akpabio

Department of Petroleum Engineering, University of Uyo, Nigeria.

Udoinyang Godwin Inyang
Department of Computer Science, University of Uyo, Nigeria and TETFund Centre of Excellence in Computational Intelligence Research, University of Uyo, Nigeria.

 

Anietie Emmanuel John
Department of Mathematics and Computer Science, Ritman University, Nigeria.

 

Mfon Okpu Esang
Department of Computer Science, Federal Polytechnic Ukana, Nigeria.

 

Eduediuyai Ekerette Dan
Department of Computer Engineering, Federal Polytechnic Ukana, Nigeria.

 

Imaobong Okpongette Akpan
Department of Mechanical Engineering, Federal Polytechnic Ukana, Nigeria.

 

Aniefiok Bassey
Department of Computer Science, Alvan Ikoku Federal University of Education, Nigeria

 

Kitoye Ebire Okonny
ICT Centre, Ignatius Ajuru University of Education, Port Harcourt, Nigeria.

Ifeanyi Bardi
Department of Computer Science, Ignatius Ajuru University of Education, Port Harcourt, Nigeria.

 

Please see the book here:- https://doi.org/10.9734/bpi/caert/v8/1874