Smart shale gas production performance analysis using machine learning applications

Abstract With the advancement of technology and innovation in the oil and gas industry, the production of liquid and gaseous hydrocarbon from conventional and unconventional resources has seen exponential growth. Recently, the USA and other oil giants have shifted their paradigm from conventional to unconventional resources of exploration and production of hydrocarbon. However, there is still a perpetual force that exists to develop and devise new innovative approaches and methodologies for the exploration, extraction efficiencies, and production performance of hydrocarbons. To better evaluate the impact of well attributes, reservoir characteristics and production behavior of well machine learning and artificial intelligence-based models have been developed by researchers that with the help of simulation and modeling provide us the true picture of reservoir performance without exploring and investing billions of dollars. This review paper encompasses the literature published in the recent years and narrated the recent development made by researchers especially in the field of production performance estimation of shale gas by developing machine learning-based models. More specifically, this paper deals with the major shale gas reservoir of North America including Marcellus shale, Eagle Ford shale, and Bakken Shale. Additionally, equations, input parameters, and formations that are considered key parameters for the development of the smart shale gas models are also discussed in this manuscript. In addition, the methodology comparison of different machine learning algorithms including their limitations and advantages are also presented.

[1]  A. R. Ismail,et al.  Analysis on the effect of different fracture geometries on the productivity of tight gas reservoirs , 2020 .

[2]  Christine Ehlig-Economides,et al.  Production-Strategy Insights Using Machine Learning: Application for Bakken Shale , 2019, SPE Reservoir Evaluation & Engineering.

[3]  Qiang Wang,et al.  Integrating linear and nonlinear forecasting techniques based on grey theory and artificial intelligence to forecast shale gas monthly production in Pennsylvania and Texas of the United States , 2019, Energy.

[4]  Alireza Shahkarami,et al.  Application of Machine Learning Algorithms for Optimizing Future Production in Marcellus Shale, Case Study of Southwestern Pennsylvania , 2018, Day 4 Wed, October 10, 2018.

[5]  Timothy R. Carr,et al.  Application of predictive data analytics to model daily hydrocarbon production using petrophysical, geomechanical, fiber-optic, completions, and surface data: A case study from the Marcellus Shale, North America , 2019, Journal of Petroleum Science and Engineering.

[6]  Fahad I. Syed,et al.  Application of ML & AI to model petrophysical and geo-mechanical properties of shale reservoirs – A systematic literature review , 2020 .

[7]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[8]  J. Friedman.  Stochastic gradient boosting , 2002 .

[9]  Rosaria Silipo,et al.  Artificial neural networks for automatic ECG analysis , 1998, IEEE Trans. Signal Process..

[10]  Vijay Kotu,et al.  Chapter 4 – Classification , 2015 .

[12]  Heng Li,et al.  Optimal Parametric Design for Water-Alternating-Gas (WAG) Process in a CO2-Miscible Flooding Reservoir , 2009 .

[13]  Marko Maucec,et al.  Application of Automated Machine Learning for Multi-Variate Prediction of Well Production , 2019, Day 3 Wed, March 20, 2019.

[14]  C. Liang,et al.  Basic characteristics and evaluation of shale oil reservoirs , 2016 .

[15]  Sathish Sankaran,et al.  Application of Machine Learning for Production Forecasting for Unconventional Resources , 2019, Proceedings of the 7th Unconventional Resources Technology Conference.

[16]  Jung,et al.  Comparative Study on Supervised Learning Models for Productivity Forecasting of Shale Reservoirs Based on a Data-Driven Approach , 2020, Applied Sciences.

[17]  Hoss Belyadi,et al.  Data-Based Smart Model for Real Time Liquid Loading Diagnostics in Marcellus Shale via Machine Learning , 2018 .

[18]  APPLICATION OF MACHINE LEARNING MODELS IN PREDICTING INITIAL GAS PRODUCTION RATE FROM TIGHT GAS RESERVOIRS , 2019, Rudarsko-geološko-naftni zbornik.

[19]  Attila Frigyesi,et al.  Artificial neural networks improve and simplify intensive care mortality prognostication: a national cohort study of 217,289 first-time intensive care unit admissions , 2019, Journal of Intensive Care.

[20]  Lulu Liao,et al.  A Big Data Study: Correlations Between EUR and Petrophysics/Engineering/Production Parameters in Shale Formations by Data Regression and Interpolation Analysis , 2019, Day 2 Wed, February 06, 2019.

[21]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[22]  Y. Liang,et al.  A Machine Learning Analysis Based on Big Data for Eagle Ford Shale Formation , 2019, Day 2 Tue, October 01, 2019.

[23]  Fahad I. Syed,et al.  Artificial lift system optimization using machine learning applications , 2020 .

[24]  Vladimir Vapnik,et al.  Support-vector networks , 2004, Machine Learning.

[25]  Husam H. Alkinani,et al.  Production Performance Estimation from Stimulation and Completion Parameters Using Machine Learning Approach in the Marcellus Shale , 2019 .

[26]  Christine Ehlig-Economides,et al.  Production Optimization Using Machine Learning in Bakken Shale , 2018 .