Supervised machine learning application of lithofacies classification for a hydrodynamically complex gas - condensate reservoir in Nam Con Son basin
Abstract
Conventional integration of rock physics and seismic inversion can quantitatively evaluate and contrast reservoir properties. However, the available output attributes are occasionally not a perfect indicator for specific information such as lithology or fluid saturation due to technology constraints. Each attribute commonly exhibits a combination of geological characteristics that could lead to subjective interpretations and provides only qualitative results. Meanwhile, machine learning (ML) is emerging as an independent interpreter to synthesise all parameters simultaneously, mitigate the uncertainty of biased cut-off, and objectively classify lithofacies on the accuracy scale.
In this paper, multiple classification algorithms including support vector machine (SVM), random forest (RF), decision tree (DT), K-nearest neighbours (KNN), logistic regression, Gaussian, Bernoulli, multinomial Naïve Bayes, and linear discriminant analysis were executed on the seismic attributes for lithofacies prediction. Initially, all data points of five seismic attributes of acoustic impedance, Lambda-Rho, Mu-Rho, density (ρ), and compressional wave to shear wave velocity (VpVs) within 25-metre radius and 25-metre interval offset top and base of reservoir were orbitally extracted on 4 wells to create the datasets. Cross-validation and grid search were also implemented on the best four algorithms to optimise the hyper-parameters for each algorithm and avoid overfitting during training. Finally, confusion matrix and accuracy scores were exploited to determine the ultimate model for discrete lithofacies prediction. The machine learning models were applied to predict lithofacies for a complex reservoir in an area of 163 km2.
From the perspective of classification, the random forest method achieved the highest accuracy score of 0.907 compared to support vector machine (0.896), K-nearest neighbours (0.895), and decision tree (0.892). At well locations, the correlation factor was excellent with 0.88 for random forest results versus sand thickness. In terms of sand and shale distribution, the machine learning outputs demonstrated geologically reasonable results, even in undrilled regions and reservoir boundary areas.
References
Phạm Hoàng Duy, Hoàng Kỳ Sơn, Trần Ngọc Thế Hùng, và Trần Vũ Tùng, “Kết quả đo độ thấm bằng nhiều phương pháp khác nhau cho vỉa turbidite mỏ Hải Thạch, bể Nam Côn Sơn”, Tạp chí Dầu khí, Số 2, trang 35 - 44, 2019.
Pham Hoang Duy, Hoang Ky Son, Trinh Xuan Vinh, and Tran Vu Tung, “Condensate banking characterization and quantification of improvement from different mitigations using pressure transient analysis: A case study in Hai Thach field offshore Vietnam”, Offshore Technology Conference Asia, Kuala Lumpur, Malaysia, 2 - 6 November 2020. DOI: 10.4043/30142-MS.
Hoang Ky Son, Tran Vu Tung, Nguyen Ngoc Tan, Truong Tu Anh, Pham Hoang Duy, Tran Ngoc Trung, Trinh Xuan Vinh, and Ngo Tuan Anh, “Successful application of machine learning to improve dynamic modeling and history matching for complex gas-condensate reservoirs in Hai Thach field, Nam Con Son basin, offshore Vietnam”, SPE Symposium: Artificial intelligence - Towards a Resilient and Efficient Energy Industry, 18 - 19 October 2021. DOI: 10.2118/208657-MS.
Hoang Ky Son, Tran Vu Tung, Nguyen Ngoc Tan, Truong Anh Tu, Pham Hoang Duy, Tran Ngoc Trung, Trinh Xuan Vinh, and Ngo Tuan Anh, “Successful case study of machine learning application to streamline and improve history matching process for complex gas-condensate reservoirs in Hai Thach field, offshore Vietnam”, SPE Middle East Oil & Gas Show and Conference, 29 November 2021. DOI: 10.2118/204835-MS.
Leila Aliouane, Sid-Ali Ouadfeul, Nouredine Djarfour, and Amar Boudella, “Lithofacies prediction from well logs data using different neural network models”, Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods (PRG-2013), 2013. DOI: 10.5220/0004380707020706.
Paolo Bestagini, Vincenzo Lipari, and Stefano Tubaro, “A machine learning approach to facies classification using well logs”, SEG Technical Program Expanded Abstracts 2017. DOI: 10.1190/segam2017-17729805.1.
Jing Jing Liu and Jian Chao Liu, “Integrating deep learning and logging data analytics for lithofacies classification and 3D modeling of tight sandstone reservoirs”, Geoscience Frontiers, Vol. 17, No. 1, 2022. DOI: 10.1016/j.gsf.2021.101311.
Randall S. Miller, Skip Rhodes, Deepak Khosla, and Fernando Nino, “Application of artificial intelligence for depositional facies recognition - Permian basin”, Unconventional Resources Technology Conference, Denver, Colorado, USA, 22 - 24 July 2019. DOI: 10.15530/urtec-2019-193.
Tran Vu Tung, Ngo Huu Hai, Hoang Ky Son, Tran Ngoc The Hung, and Joseph J. Lambiase, “Depositional facies prediction using artificial intelligence to improve reservoir characterization in a mature field of Nam Con Son basin, offshore Vietnam”, Offshore Technology Conference Asia, Kuala Lumpur, Malaysia, 2 - 6 November 2020. DOI: 10.4043/30086-MS.
Hiren Maniar, Srikanth Ryali, Mandar S. Kulkarni, and Aria Abubakar, “Machine learning methods in geoscience”, SEG Technical Program Expanded Abstracts 2018. DOI: 10.1190/segam2018-2997218.1.
Seth Brazell, Alex Bayeh, Michael Ashby, and Darrin Burton, “A machine-learning-based approach to assistive well-log correlation”, Petrophysics, Vol. 60, No. 4, pp. 469 - 479, 2019. DOI: 10.30632/PJV60N4-2019a1.
Satinder Chopra and Kurt J. Marfurt, “Seismic facies characterization using some unsupervised machine learning methods”, SEG Technical Program Expanded Abstracts 2018. DOI: 10.1190/segam2018-2997356.1.
Vladimir Puzyrev and Chris Elders, “Unsupervised seismic facies classification using deep convolutional autoencoder”, Geophysics, Vol. 87, No. 4, 2022. DOI: 10.1190/geo2021-0016.1.
Thilo Wrona, Indranil Pan, Robert L. Gawthorpe, and Haakon Fossen, “Seismic facies analysis using machine learning”, Geophysics, Vol. 83, No. 5, 2018. DOI: 10.1190/geo2017-0595.1.
1. The Author assigns all copyright in and to the article (the Work) to the Petrovietnam Journal, including the right to publish, republish, transmit, sell and distribute the Work in whole or in part in electronic and print editions of the Journal, in all media of expression now known or later developed.
2. By this assignment of copyright to the Petrovietnam Journal, reproduction, posting, transmission, distribution or other use of the Work in whole or in part in any medium by the Author requires a full citation to the Journal, suitable in form and content as follows: title of article, authors’ names, journal title, volume, issue, year, copyright owner as specified in the Journal, DOI number. Links to the final article published on the website of the Journal are encouraged.
