پیش‌بینی فشار منفذی داده‌های گمانه‌های پتروفیزیکی با بهره‌گیری از روش‌های مبتنی بر هوش مصنوعی

نوع مقاله : علمی - پژوهشی

نویسندگان

1 استادیار گروه مهندسی معدن، دانشگاه صنعتی بیرجند، بیرجند، ایران

2 باشگاه پژوهشگران جوان و نخبگان ، واحد اهواز، دانشگاه آزاد اسلامی، اهواز، ایران

3 کارشناسی ارشد زمین شناسی، دانشکده علوم زمین، دانشگاه تحصیلات تکمیلی علوم پایه زنجان، ایران

چکیده

فشار منفذی یکی از مهم‌ترین پارامترهای مخزنی-حفاری به شمار می‌رود و آگاهی از این فشار، برای کاهش هزینه‌های حفاری، افزایش ایمنی چاه و پیشگیری از خطرات احتمالی ضروری می باشد. تحقیقات نشان داده است که معادلات تجربی تنها برای نواحی خاصی دقت مطلوبی دارند زیرا، اکثر معادلات تجربی بر اساس مجموعه داده‌های محدود، جمع‌آوری‌ و توسعه‌ یافته اند. بنابراین، روش‌های هوشمند جای خود را به این معادلات داده اند. مطالعه پیش رو، از 2827 داده ست مربوط به سه چاه از میادین نفتی واقع در جنوب غرب ایران بهره برده است. متغیرهای ورودی مورد استفاده به‌منظور پیش‌بینی فشار منفذی شامل 9 متغیر بوده که با استفاده از روش انتخاب ویژگی، برگزیده ‌شده‌اند. در این مطالعه از چهار الگوریتم هوش مصنوعی شامل الگوریتم جنگل تصادفی ، الگوریتم رگرسیون بردار پشتیبان ، الگوریتم شبکه عصبی مصنوعی و الگوریتم درخت تصمیم به‌منظور پیش‌بینی فشار منفذی استفاده‌ شده است. پس از بررسی نتایج مشخص گردید که دقت عملکرد الگوریتم درخت تصمیم، بیشتر از سه الگوریتم دیگر می باشد به طوری که برای این الگوریتم مقدار مربع ضریب همبستگی (R^2) 9985/0 و مجذور میانگین مربعات خطا (RMSE) 460/14psi به دست آمده است. از جمله مزیت‌های این الگوریتم، ارائه بهترین نتیجه بدون نیاز به دانش آماری، جدا کردن داده‌های غیرضروری، آماده‌سازی داده‌ها در زمان کوتاه و کاهش خطای نسبی با یافتن گره اصلی تصمیم گیر می باشد. بنابراین می‌توان چنین نتیجه گرفت که با توسعه این تکنیک، برای تعداد داده‌های کم از هر میدان، دقت عملکرد بالایی نتیجه می شود.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Comparison of artificial intelligence algorithms to predict pore pressure using petrophysical log data

نویسندگان [English]

  • Meysam Rajabi 1
  • Hamzeh Ghorbani 2
  • Sahar Lajmorak 3
1 Assistant Professor, Department of Mining Engineering, Birjand University of Technology, Birjand, Iran
2 Young Researchers and Elite Club, Ahvaz Branch, Islamic Azad University, Ahvaz, Iran
3 Master of Geology, Faculty of Earth Sciences, Zanjan University of Basic Sciences, Iran
چکیده [English]

Pore pressure is one of the most important reservoir-drilling parameters and knowledge of this pressure is essential for drilling costs, well safety and prevention of potential hazards. Research has shown that experimental equations have good performance accuracy only for certain regions. Most of these experimental equations have been compiled and developed based on a limited data set. Therefore, these correlations are valid in the range of changes in the parameters of those fields and are not valid for other areas. Therefore, artificial intelligent methods have given way to empirical equations. In this study, 2827 data related to three wells from one of the oil fields located in the southwest of Iran have been used. The input variables used in this paper to predict the pore pressure include 9 variables that have been selected using the feature selection method. In this study, 4 artificial intelligence algorithms include; random forest algorithm, support vector regression algorithm, artificial neural network algorithm and decision tree algorithm have been used to predict the pore pressure. After reviewing the results, it was found that the performance accuracy of the decision tree algorithm is higher than the other three algorithms (performance accuracy for the entire data set including R2 = 0.9985 and RMSE = 14.460 psi). Among the advantages of this algorithm compared to other algorithms are the best results without the need for statistical knowledge, separation of unnecessary data, short time to prepare data and reduction of relative error by finding the main node of the decision maker and analyzing it. Therefore, it can be concluded that with the development of this technique, it is possible to have high performance accuracy for a small amount of data from each field.

کلیدواژه‌ها [English]

  • Pore pressure
  • Artificial intelligence algorithms
  • Petrophysical data
  • Decision tree algorithm
  • Feature selection
[1]          Hu L, Deng J, Zhu H, Lin H, Chen Z, Deng F, et al. A new pore pressure prediction method-back propagation artificial neural network. Electron J Geotech Eng 2013;18:4093-107. doi: http://www.ejge.com/2013/Ppr2013.371mlr.pdf.
[2]          Yu H, Chen G, Gu H. A machine learning methodology for multivariate pore-pressure prediction. Computers & Geosciences 2020;143:104548. doi: https://doi.org/10.1016/j.cageo.2020.104548.
[3]          Bahmaei Z, Hosseini E. Pore pressure prediction using seismic velocity modeling: case study, Sefid-Zakhor gas field in Southern Iran. Journal of Petroleum Exploration and Production Technology 2020;10(3):1051-62. doi: https://link.springer.com/article/10.1007/s13202-019-00818-y.
[4]          Zhang Y, Lv D, Wang Y, Liu H, Song G, Gao J. Geological characteristics and abnormal pore pressure prediction in shale oil formations of the Dongying depression, China. Energy Science & Engineering 2020;8(6):1962-79. doi: https://doi.org/10.1002/ese3.641.
[5]          Mousavipour F, Riahi MA, Moghanloo HGJJoPE, Technology P. Prediction of in situ stresses, mud window and overpressure zone using well logs in South Pars field. 2020;10(5):1869-79. doi: https://doi.org/10.2118/189665-PA.
[6]          Ahedor MK-N, Anumah P, Sarkodie-Kyeremeh J. Post-Drill Pore Pressure and Fracture Gradient Analyses of Y-Field in the Offshore Tano Basin of Ghana. OnePetro; 2020. doi: https://doi.org/10.2118/203659-MS.
[7]          Mahetaji M, Brahma J, Sircar A. Pre-drill pore pressure prediction and safe well design on the top of Tulamura anticline, Tripura, India: a comparative study. Journal of Petroleum Exploration and Production Technology 2020;10(3):1021-49. doi: https://link.springer.com/article/10.1007/s13202-019-00816-0.
[8]          Oloruntobi O, Falugba O, Ekanem-Attah O, Awa C, Butt S. The Niger Delta basin fracture pressure prediction. Environmental Earth Sciences 2020;79(13):1-11. doi: https://link.springer.com/article/10.1007/s12665-020-09081-5.
[9]          Darvishpour A, Seifabad MC, Wood DA, Ghorbani H. Wellbore stability analysis to determine the safe mud weight window for sandstone layers. Petroleum Exploration and Development 2019;46(5):1031-8. doi: https://doi.org/10.1016/S1876-3804(19)60260-0.
[10]        Li S, George J, Purdy C. Pore-pressure and wellbore-stability prediction to increase drilling efficiency. Journal of Petroleum Technology 2012;64(02):98-101. doi: https://doi.org/10.2118/144717-JPT.
[11]        Benemaran RS, Esmaeili-Falak M. Optimization of cost and mechanical properties of concrete with admixtures using MARS and PSO. Computers and Concrete, An International Journal 2020;26(4):309-16. doi: https://doi.org/10.1061/(ASCE)CR.1943-5495.0000188.
[12]        Zhu W, Huang L, Mao L, Esmaeili‐Falak M. Predicting the uniaxial compressive strength of oil palm shell lightweight aggregate concrete using artificial intelligence‐based algorithms. Structural Concrete 2021. doi: https://doi.org/10.1002/suco.202100656.
[13]        Hottmann CE, Johnson RK. Estimation of formation pressures from log-derived shale properties. Journal of Petroleum Technology 1965;17(06):717-22. doi: https://doi.org/10.2118/1110-PA.
[14]        Bingham G. A new approach to interpreting rock drillability. TECHNICAL MANUAL REPRINT, OIL AND GAS JOURNAL, 1965 93 P 1965. doi:
[15]        Jorden JR, Shirley OJ. Application of drilling performance data to overpressure detection. Journal of Petroleum Technology 1966;18(11):1387-94. doi: https://doi.org/10.2118/1407-PA.
[16]        Terzaghi K, Peck RB, Mesri G. Soil mechanics in engineering practice. John Wiley & Sons; 1996.
[17]        Biot MA, Willis DG. The elastic coefficients of the theory of consolidation. 1957. doi: https://doi.org/10.1115/1.4011606.
[18]        Eaton BA. The equation for geopressure prediction from well logs. OnePetro; 1975. doi: https://doi.org/10.2118/5544-MS.
[19]        Bowers GL. Pore pressure estimation from velocity data: Accounting for overpressure mechanisms besides undercompaction. SPE Drilling & Completion 1995;10(02):89-95. doi: https://doi.org/10.2118/27488-PA.
[20]        Yoshida C, Ikeda S, Eaton BA. An investigative study of recent technologies used for prediction, detection, and evaluation of abnormal formation pressure and fracture pressure in North and South America. OnePetro; 1996. doi: https://doi.org/10.2118/36381-MS.
[21]        Shen Y, Luan G, Zhang H, Liu Q, Zhang J, Ge H. Novel method for calculating the effective stress coefficient in a tight sandstone reservoir. KSCE Journal of Civil Engineering 2017;21(6):2467. doi: 10.1007/s12205-016-0514-5.
[22]        Farsi M, Mohamadian N, Ghorbani H, Wood DA, Davoodi S, Moghadasi J, et al. Predicting Formation Pore-Pressure from Well-Log Data with Hybrid Machine-Learning Optimization Algorithms. Natural Resources Research 2021:1-27. doi: https://link.springer.com/article/10.1007/s11053-021-09852-2.
[23]        Abad ARB, Mousavi S, Mohamadian N, Wood DA, Ghorbani H, Davoodi S, et al. Hybrid machine learning algorithms to predict condensate viscosity in the near wellbore regions of gas condensate reservoirs. Journal of Natural Gas Science and Engineering 2021;95:104210. doi: https://doi.org/10.1016/j.jngse.2021.104210.
[24]        Naveshki M, Naghiei A, Soltani Tehrani P, Ahmadi Alvar M, Ghorbani H, Mohamadian N, et al. Prediction of bubble point pressure using new hybrid computationail intelligence models. Journal of Chemical and Petroleum Engineering 2021. doi: 10.22059/JCHPE.2021.314719.1341.
[25]        Hazbeh O, Ahmadi Alvar M, Aghdam K-y, Ghorbani H, Mohamadian N, Moghadasi J. Hybrid computing models to predict oil formation volume factor using multilayer perceptron algorithm. Journal of Petroleum and Mining Engineering 2021:14-27. doi: 10.21608/JPME.2021.52149.1062.
[26]        Wang Y, Ma H, Fu W. Formation pressure prediction based on hybrid genetic algorithm. IEEE; 2010:2535-8. doi: 10.1109/ICOSP.2010.5656925.
[27]        Abidin MH. Pore Pressure Estimation Using Artificial Neural Network. 2014. doi: http://utpedia.utp.edu.my/id/eprint/14317.
[28]        Haris A, Sitorus RJ, Riyanto A. Pore pressure prediction using probabilistic neural network: case study of South Sumatra Basin. IOP Conference Series: Earth and Environmental Science 2017;62:012021. doi: 10.1088/1755-1315/62/1/012021.
[29]        Rashidi M, Asadi A. An Artificial Intelligence Approach in Estimation of Formation Pore Pressure by Critical Drilling Data. OnePetro; 2018. doi:
[30]        Karmakar M, Maiti S. Short term memory efficient pore pressure prediction via Bayesian neural networks at Bering Sea slope of IODP expedition 323. Measurement 2019;135:852-68. doi: https://onepetro.org/ARMAUSRMS/proceedings-abstract/ARMA18/All-ARMA18/ARMA-2018-1098/124075.
[31]        Abdelaal A, Elkatatny S, Abdulraheem A. Data-Driven Modeling Approach for Pore Pressure Gradient Prediction while Drilling from Drilling Parameters. ACS omega 2021. doi: https://doi.org/10.1021/acsomega.1c01340.
[32]        Legg S, Hutter M. A collection of definitions of intelligence. Frontiers in Artificial Intelligence and applications 2007;157:17. doi:
[33]        Russell S, Norvig P. Artificial intelligence: a modern approach. 2002. doi: https://storage.googleapis.com/pub-tools-public-publication-data/pdf/27702.pdf.
[34]        Poole D, Mackworth A, Goebel R. Computational Intelligence. 1998. doi:
[35]        Nie G, Rowe W, Zhang L, Tian Y, Shi Y. Credit card churn forecasting by logistic regression and decision tree. Expert Systems with Applications 2011;38(12):15273-85. doi: https://doi.org/10.1016/j.eswa.2011.06.028.
[36]        Tsai C-F, Chiou Y-J. Earnings management prediction: A pilot study of combining neural networks and decision trees. Expert systems with applications 2009;36(3):7183-91. doi: https://doi.org/10.1016/j.eswa.2008.09.025.
[37]        Lorena AC, de Carvalho AC. Protein cellular localization prediction with support vector machines and decision trees. Computers in Biology and Medicine 2007;37(2):115-25. doi: https://doi.org/10.1016/j.compbiomed.2006.01.003.
[38]        Osei-Bryson K-M. Evaluation of decision trees: a multi-criteria approach. Computers & Operations Research 2004;31(11):1933-45. doi: https://doi.org/10.1016/S0305-0548(03)00156-4.
[39]        Fakhari A, Moghadam AME. Combination of classification and regression in decision tree for multi-labeling image annotation and retrieval. Applied Soft Computing 2013;13(2):1292-302. doi: https://doi.org/10.1016/j.asoc.2012.10.019.
[40]        Liu J, Sui C, Deng D, Wang J, Feng B, Liu W, et al. Representing conditional preference by boosted regression trees for recommendation. Information Sciences 2016;327:1-20. doi: https://doi.org/10.1016/j.ins.2015.08.001.
[41]        Ahmad MS, Adnan SM, Zaidi S, Bhargava P. A novel support vector regression (SVR) model for the prediction of splice strength of the unconfined beam specimens. Construction and Building Materials 2020;248:118475. doi: https://doi.org/10.1016/j.conbuildmat.2020.118475.
[42]        Rui J, Zhang H, Zhang D, Han F, Guo Q. Total organic carbon content prediction based on support-vector-regression machine with particle swarm optimization. Journal of Petroleum Science and Engineering 2019;180:699-706. doi: https://doi.org/10.1016/j.petrol.2019.06.014.
[43]        Shao M, Wang X, Bu Z, Chen X, Wang Y. Prediction of energy consumption in hotel buildings via support vector machines. Sustainable Cities and Society 2020;57:102128. doi: https://doi.org/10.1016/j.scs.2020.102128.
[44]        Smola AJ, Schölkopf B. A tutorial on support vector regression. Statistics and computing 2004;14(3):199-222. doi: https://link.springer.com/article/10.1023/B:STCO.0000035301.49549.88.
[45]        Vapnik V. The nature of statistical learning theory. Springer science & business media; 2013.
[46]        Hashemitaheri M, Mekarthy SMR, Cherukuri H. Prediction of specific cutting forces and maximum tool temperatures in orthogonal machining by support vector and Gaussian process regression methods. Procedia Manufacturing 2020;48:1000-8. doi: https://doi.org/10.1016/j.promfg.2020.05.139.
[47]        Zhou X, Lu P, Zheng Z, Tolliver D, Keramati A. Accident prediction accuracy assessment for highway-rail grade crossings using random forest algorithm compared with decision tree. Reliability Engineering & System Safety 2020;200:106931. doi: https://doi.org/10.1016/j.ress.2020.106931.
[48]        Grape S, Branger E, Elter Z, Balkeståhl LP. Determination of spent nuclear fuel parameters using modelled signatures from non-destructive assay and Random Forest regression. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 2020;969:163979. doi: https://doi.org/10.1016/j.nima.2020.163979.
[49]        Ahmad MW, Reynolds J, Rezgui Y. Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees. Journal of cleaner production 2018;203:810-21. doi: https://doi.org/10.1016/j.jclepro.2018.08.207.
[50]        Ahmad MW, Mourshed M, Rezgui Y. Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption. Energy and Buildings 2017;147:77-89. doi: https://doi.org/10.1016/j.enbuild.2017.04.038.
[51]        Shahbaz M, Taqvi SA, Loy ACM, Inayat A, Uddin F, Bokhari A, et al. Artificial neural network approach for the steam gasification of palm oil waste using bottom ash and CaO. Renewable Energy 2019;132:243-54. doi: https://doi.org/10.1016/j.renene.2018.07.142.
[52]        Kartal F, Özveren U. A deep learning approach for prediction of syngas lower heating value from CFB gasifier in Aspen plus®. Energy 2020;209:118457. doi: https://doi.org/10.1016/j.energy.2020.118457.
[53]        Abad ARB, Ghorbani H, Mohamadian N, Davoodi S, Mehrad M, Aghdam SK-y, et al. Robust hybrid machine learning algorithms for gas flow rates prediction through wellhead chokes in gas condensate fields. Fuel 2022;308:121872. doi: https://doi.org/10.1016/j.fuel.2021.121872.
[54]        Wahab MNA, Nefti-Meziani S, Atyabi A. A comprehensive review of swarm optimization algorithms. PloS one 2015;10(5):1-36. doi: https://doi.org/10.1371/journal.pone.0122827.
[55]        Chandrashekar G, Sahin F. A survey on feature selection methods. Computers and Electrical Engineering 2014;40(1):16-28. doi: https://doi.org/10.1016/j.compeleceng.2013.11.024.
[56]        Al-Mudhafar WJ. Bayesian and LASSO regressions for comparative permeability modeling of sandstone reservoirs. Natural Resources Research 2019;28(1):47-62. doi: https://link.springer.com/article/10.1007/s11053-018-9370-y.
[57]        Jain A, Zongker D. Feature selection: Evaluation, application, and small sample performance. IEEE transactions on pattern analysis and machine intelligence 1997;19(2):153-8. doi: https://doi.org/10.1109/34.574797.
[58]        Fu G-H, Wu Y-J, Zong M-J, Pan J. Hellinger distance-based stable sparse feature selection for high-dimensional class-imbalanced data. BMC bioinformatics 2020;21(1):1-14. doi: https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-020-3411-3.
[59]        Salehi M, Farhadi S, Moieni A, Safaie N, Ahmadi HJFips. Mathematical modeling of growth and paclitaxel biosynthesis in Corylus avellana cell culture responding to fungal elicitors using multilayer perceptron-genetic algorithm. 2020;11. doi: 10.3389/fpls.2020.01148.
[60]        Jotheeswaran J, Koteeswaran S. Sentiment Polarity Classification Using Conjure of Genetic Algorithm and Differential Evolution Methods for Optimized Feature Selection. Recent Advances in Computer Science and Communications (Formerly: Recent Patents on Computer Science) 2020;13(6):1284-91. doi: https://doi.org/10.2174/2213275911666180904110105.