Predicting shear wave velocity from conventional well logs with deep and hybrid machine learning algorithms

Xehetasun bibliografikoak
Parent link:Journal of Petroleum Exploration and Production
Vol. 13, iss. 1.— 2023.— [P. 19-42]
Erakunde egilea: Национальный исследовательский Томский политехнический университет Инженерная школа природных ресурсов Отделение нефтегазового дела
Beste egile batzuk: Rajabi M. Meysam, Hazbeh O. Omid, Davoodi Sh. Shadfar, Wood D. A. David, Tehrani P. S. Pezhman Soltani, Ghorbani H. Hamzeh, Mehrad M. Mohammad, Mohamadian N. Nima, Rukavishnikov V. S. Valery Sergeevich, Radwan A. E. Ahmed
Gaia:Title screen
Shear wave velocity (VS) data from sedimentary rock sequences is a prerequisite for implementing most mathematical models of petroleum engineering geomechanics. Extracting such data by analyzing finite reservoir rock cores is very costly and limited. The high cost of sonic dipole advanced wellbore logging service and its implementation in a few wells of a field has placed many limitations on geomechanical modeling. On the other hand, shear wave velocity VS tends to be nonlinearly related to many of its influencing variables, making empirical correlations unreliable for its prediction. Hybrid machine learning (HML) algorithms are well suited to improving predictions of such variables. Recent advances in deep learning (DL) algorithms suggest that they too should be useful for predicting VS for large gas and oil field datasets but this has yet to be verified. In this study, 6622 data records from two wells in the giant Iranian Marun oil field (MN#163 and MN#225) are used to train HML and DL algorithms. 2072 independent data records from another well (MN#179) are used to verify the VS prediction performance based on eight well-log-derived influencing variables. Input variables are standard full-set recorded parameters in conventional oil and gas well logging data available in most older wells.
DL predicts VS for the supervised validation subset with a root mean squared error (RMSE) of 0.055 km/s and coefficient of determination (R2) of 0.9729. It achieves similar prediction accuracy when applied to an unseen dataset. By comparing the VS prediction performance results, it is apparent that the DL convolutional neural network model slightly outperforms the HML algorithms tested. Both DL and HLM models substantially outperform five commonly used empirical relationships for calculating VS from Vp relationships when applied to the Marun Field dataset. Concerns regarding the model's integrity and reproducibility were also addressed by evaluating it on data from another well in the field. The findings of this study can lead to the development of knowledge of production patterns and sustainability of oil reservoirs and the prevention of enormous damage related to geomechanics through a better understanding of wellbore instability and casing collapse problems.
Режим доступа: по договору с организацией-держателем ресурса
Hizkuntza:ingelesa
Argitaratua: 2023
Gaiak:
Sarrera elektronikoa:https://doi.org/10.1007/s13202-022-01531-z
Formatua: Baliabide elektronikoa Liburu kapitulua
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=668644

MARC

LEADER 00000naa0a2200000 4500
001 668644
005 20250910102812.0
035 |a (RuTPU)RU\TPU\network\39881 
035 |a RU\TPU\network\36870 
090 |a 668644 
100 |a 20230116d2023 k||y0rusy50 ba 
101 0 |a eng 
102 |a DE 
135 |a drcn ---uucaa 
181 0 |a i  
182 0 |a b 
200 1 |a Predicting shear wave velocity from conventional well logs with deep and hybrid machine learning algorithms  |f M. Rajabi, O. Hazbeh, Sh. Davoodi [et al.] 
203 |a Text  |c electronic 
300 |a Title screen 
330 |a Shear wave velocity (VS) data from sedimentary rock sequences is a prerequisite for implementing most mathematical models of petroleum engineering geomechanics. Extracting such data by analyzing finite reservoir rock cores is very costly and limited. The high cost of sonic dipole advanced wellbore logging service and its implementation in a few wells of a field has placed many limitations on geomechanical modeling. On the other hand, shear wave velocity VS tends to be nonlinearly related to many of its influencing variables, making empirical correlations unreliable for its prediction. Hybrid machine learning (HML) algorithms are well suited to improving predictions of such variables. Recent advances in deep learning (DL) algorithms suggest that they too should be useful for predicting VS for large gas and oil field datasets but this has yet to be verified. In this study, 6622 data records from two wells in the giant Iranian Marun oil field (MN#163 and MN#225) are used to train HML and DL algorithms. 2072 independent data records from another well (MN#179) are used to verify the VS prediction performance based on eight well-log-derived influencing variables. Input variables are standard full-set recorded parameters in conventional oil and gas well logging data available in most older wells. 
330 |a DL predicts VS for the supervised validation subset with a root mean squared error (RMSE) of 0.055 km/s and coefficient of determination (R2) of 0.9729. It achieves similar prediction accuracy when applied to an unseen dataset. By comparing the VS prediction performance results, it is apparent that the DL convolutional neural network model slightly outperforms the HML algorithms tested. Both DL and HLM models substantially outperform five commonly used empirical relationships for calculating VS from Vp relationships when applied to the Marun Field dataset. Concerns regarding the model's integrity and reproducibility were also addressed by evaluating it on data from another well in the field. The findings of this study can lead to the development of knowledge of production patterns and sustainability of oil reservoirs and the prevention of enormous damage related to geomechanics through a better understanding of wellbore instability and casing collapse problems. 
333 |a Режим доступа: по договору с организацией-держателем ресурса 
461 |t Journal of Petroleum Exploration and Production 
463 |t Vol. 13, iss. 1  |v [P. 19-42]  |d 2023 
610 1 |a труды учёных ТПУ 
610 1 |a электронный ресурс 
610 1 |a shear wave velocity 
610 1 |a hybrid machine learning 
610 1 |a deep learning 
610 1 |a well-log influencing variables 
610 1 |a multi-well dataset 
610 1 |a convolutional neural network 
610 1 |a поперечные волны 
610 1 |a скорость 
610 1 |a гибридное обучение 
610 1 |a машинное обучение 
610 1 |a каротаж 
610 1 |a сверточные нейронные сети 
701 1 |a Rajabi  |b M.  |g Meysam 
701 1 |a Hazbeh  |b O.  |g Omid 
701 1 |a Davoodi  |b Sh.  |c specialist in the field of petroleum engineering  |c Research Engineer of Tomsk Polytechnic University  |f 1990-  |g Shadfar  |3 (RuTPU)RU\TPU\pers\46542  |9 22200 
701 1 |a Wood  |b D. A.  |g David 
701 1 |a Tehrani  |b P. S.  |g Pezhman Soltani 
701 1 |a Ghorbani  |b H.  |g Hamzeh 
701 1 |a Mehrad  |b M.  |g Mohammad 
701 1 |a Mohamadian  |b N.  |g Nima 
701 1 |a Rukavishnikov  |b V. S.  |c Director of the Center for Training and Retraining of Oil and Gas Specialists, Associate Professor of Tomsk Polytechnic University, Candidate of Technical Sciences  |c Engineer of Tomsk Polytechnic University  |f 1984-  |g Valery Sergeevich  |3 (RuTPU)RU\TPU\pers\34050  |9 17614 
701 1 |a Radwan  |b A. E.  |g Ahmed 
712 0 2 |a Национальный исследовательский Томский политехнический университет  |b Инженерная школа природных ресурсов  |b Отделение нефтегазового дела  |3 (RuTPU)RU\TPU\col\23546 
801 0 |a RU  |b 63413507  |c 20230329  |g PSBO 
856 4 |u https://doi.org/10.1007/s13202-022-01531-z 
942 |c CF