تعیین مشخصات جریان‌های دو فازی نفت- آب توسط شبکه عصبی کانولوشنی جریانی

نوع مقاله : مقاله پژوهشی

نویسندگان

گروه مهندسی نفت،دانشکده نفت و مهندسی شیمی، دانشگاه آزاد اسلامی، واحد علوم و تحقیقات، تهران، ایران

چکیده

تا به امروز، مدل‌های مکانیکی و روابط تجربی مختلفی برای توصیف و مدل‌سازی سیستم‌های جریان دو فازی نفت- آب معرفی شده‌اند. اما، در اکثر این مدل‌ها و روابط پیشنهادی از مفروضات ساده با رویکرد حل تکرار شونده استفاده شده، که از دقت کافی جهت تخمین خصوصیات جریانی برخوردار نمی‌باشند. هدف از مطالعه حاضر، غلبه بر این مشکل با کمک توسعه یک شبکه عصبی کانولوشنالی جریانی از طریق یادگیری عمیق می‌باشد. بدین منظور، 270 آزمایش جریانی شامل آزمایش‌های جریانی پراکنده آب در نفت، دوگانه پیوسته و پراکنده نفت در آب در دو حالت افقی و شیب دار (°30) انجام گردیده است. شبکه عصبی بر روی 70% این داده‌های آزمایشگاهی آموزش داده شد. لازم به توضیح است که از تصاویر الگوی جریانی دو بعدی به عنوان داده‌های ورودی و از الگوهای جریان و مقادیر کسر حجمی پسماند به عنوان داده‌های خروجی استفاده شده است. نتایج حاصل از این مطالعه نماینگر آن است که مدل شبکه عصبی کانولوشنالی جریانی آموزش داده شده بر روی داده‌های آزمایشگاهی قادر است رژیم‌های جریان را با دقت 91% و 96% به ترتیب در جریان‌های افقی و شیبدار پیش‌بینی نماید. این مدل همچنین قادر است کسر حجمی پسماند را با یک خطای معقول 22/1% و 98/0% به ترتیب در جریان‌های افقی و شیبدار پیش‌بینی کند. از این‌رو می‌توان گفت که رویکرد پیشنهادی قادر به پیش‌بینی خودکار و دقیق رژیم جریان و کسر حجمی پسماند در جریان‌های افقی و شیبدار از طریق تصاویر جریان است.
 

کلیدواژه‌ها


عنوان مقاله [English]

Determining Characteristics of Two-Phase Oil-Water Flows by the Convolutional Neural Network

نویسندگان [English]

  • Amir Pouria Sadegh Samimi
  • Ali Esfandyari Bayat
  • Aَbulqasem Emamzadeh
Department of Reservoir Engineering, Faculty of Petroleum and Chemical Engineering, Science and Research Branch, Islamic Azad University (IAU), Tehran, Iran
چکیده [English]

To date, various mechanistic models and empirical correlations have been developed to characterize and model two phase oil-water flow systems. However, in the most of these proposed models and correlations, simplified assumptions with the iterative solutions approach have been utilized, which do not have enough accuracy to estimate the flow characteristics. The aim of this study is to overcome this problem by developing a convolutional neural network through the deep learning. For this purpose, 270 flow tests including dispersed water-in-oil, dual continuous and dispersed oil-in-water flow tests have been conducted in the both horizontal and inclined (30o) states. The neural network was trained on 70% of the achieved laboratory data. It is necessary to explain that two-dimensional flow pattern images were used as the input data and flow patterns and liquid holdup fraction values were applied as the output data. The results of this study revealed that the applied flow convolutional neural network model is able to predict the flow regimes with 91% and 96% accuracies in the horizontal and inclined flows, respectively. This model is also able to predict the liquid holdup fraction with a reasonable error of 1.22% and 0.98% in horizontal and inclined flows, respectively. Therefore, it can be concluded that the proposed approach is able to automatically and accurately predict the flow regimes and liquid holdup fractions through flow images in the both horizontal and inclined states.
 

کلیدواژه‌ها [English]

  • Flow Convolutional Neural Network (FCNN)
  • Oil-Water Two-phase Flow
  • Deep Learning
  • Image Processing
[1]. Beggs D H, Brill J P (1973) A study of two-phase flow in inclined pipes, Journal of Petroleum Technology, 25: 607–617, doi.org/10.2118/4007-PA.##
[2]. Carcione J M, Picotti S, Santos J E, Qadrouh A, Almalki H S (2014) Numerical simulation of two-phase fluid flow, Journal of Petroleum Exploration and Production Technology, 4: 233–243. ##
[3]. Trallero J L, Sarica C, Brill J P (1997) A study of oil/water flow patterns in horizontal pipes, SPE Production and Facilities,12, 03: 165-172, doi.org/10.2118/36609-PA. ##
[4]. Kim H J, Kim G N, Kim Y J, Woo N S, Huh S C (2021) A study on the separation efficiency of in-line type subsea oil-water separator, Journal of the Korean Society of Industry Convergence, 24: 253–260, doi.org/10.21289/KSIC.2021.24.3.253. ##
[5]. Hussein M M, Al-Sarkhi A, Badr H M, Habib M A (2019) CFD modeling of liquid film reversal of two-phase flow in vertical pipes, Journal of Petroleum Exploration and Production Technology, 9: 3039-3070. ##
[6]. Shams R, Tavakoli A, Shad S (2017) Experimental investigation of two-phase flow in horizontal wells: Flow regime assessment and pressure drop analysis, Experimental Thermal and Fluid Science, 88: 55-64, doi.org/10.1016/j.expthermflusci.2017.05.011. ##
[7]. Li Z C, Fan C L (2020) A novel method to identify the flow pattern of oil–water two-phase flow, Journal of Petroleum Exploration and Production Technology, 10: 3723-3732. ##
[8]. Kabiri Samani A R, Borghei S M (2010) Pressure loss in a horizontal two-phase slug flow, Journal of Fluids Engineering, 132: 7, doi.org/10.1115/1.4001969. ##
[9]. Abubakar A, Al-Wahaibi Y, Al-Wahaibi T, Al-Hashmi A R, Al-Ajmi A, Eshrati M (2018) Effect of pipe diameter on horizontal oil-water flow before and after addition of drag-reducing polymer part II: holdup and slip ratio, Journal of Petroleum Science and Engineering, 162: 143-149, doi.org/10.1016/j.petrol.2017.12.015. ##
[10]. Angeli P. Hewitt GF (2000) Flow structure in horizontal oil--water flow, International Journal of Multiphase Flow, 26: 1117–1140, doi.org/10.1016/S0301-9322(99)00081-6. ##
[11]. Baik S, Hanratty T J (2003) Effects of a drag reducing polymer on stratified gas–liquid flow in a large diameter horizontal pipe, International Journal of Multiphase Flow, 29, 11: 1749-1757, doi.org/10.1016/j.ijmultiphaseflow.2003.07.004. ##
[12]. Burlutskii E (2018) CFD study of oil-in-water two-phase flow in horizontal and vertical pipes, Journal of Petroleum Science and Engineering, 162: 524-531, doi.org/10.1016/j.petrol.2017.10.035. ##
[13]. Charles M E, Govier G T, Hodgson G W (1961) The horizontal pipeline flow of equal density oil‐water mixtures, The Canadian Journal of Chemical Engineering, 39, 1: 27-36, doi.org/10.1002/cjce.5450390106. ##
[14]. Edomwonyi-Otu L C, Angeli P (2015) Pressure drop and holdup predictions in horizontal oil–water flows for curved and wavy interfaces, Chemical Engineering Research and Design, 93: 55-65, doi.org/10.1016/j.cherd.2014.06.009. ##
[15]. Hanafizadeh P, Hojati A, Karimi A (2015) Experimental investigation of oil–water two phase flow regime in an inclined pipe, Journal of Petroleum Science and Engineering, 136: 12-22, doi.org/10.1016/j.petrol.2015.10.031. ##
[16]. Zhu Y, Wu X, Zhao R (2017) R32 flow boiling in horizontal mini channels: Part I. Two-phase flow patterns, International Journal of Heat and Mass Transfer, 115: 1223-1232, doi.org/10.1016/j.ijheatmasstransfer.2017.07.101. ##
[17] . میثاق ن، نیسانی سامانی ن, عبدالهی کاکرودی ع، علوی پناه س ک، بحرودی ع (1396) مدل‌سازی پهنه‌های اکتشاف نفتی با شبکه عصبی پرسپترون چند لایه (MLP) در GIS. پژوهش نفت، 26، (6-95): 160-148، doi: 10.22078/pr.2017.724. ##
[18]. Bonizzi M, Issa R I (2003) A model for simulating gas bubble entrainment in two-phase horizontal slug flow, International Journal of Multiphase Flow, 29, 11: 1685-1717, doi.org/10.1016/j.ijmultiphaseflow.2003.09.001. ##
[19]. Al-Wahaibi T, Smith M, Angeli P (2007) Effect of drag-reducing polymers on horizontal oil--water flows, Journal of Petroleum Science and Engineering, 57, 3-4: 334–346, doi.org/10.1016/j.petrol.2006.11.002. ##
[20]. Abubakar A, Al-Wahaibi T, Al-Hashmi A R, Al-Wahaibi Y, Al-Ajmi A, Eshrati M (2015) Influence of drag-reducing polymer on flow patterns, drag reduction and slip velocity ratio of oil–water flow in horizontal pipe, International Journal of Multiphase Flow, 73: 1-10, doi.org/10.1016/j.ijmultiphaseflow.2015.02.016. ##
[21]. Wyatt N B, Gunther C M, Liberatore M W (2011) Drag reduction effectiveness of dilute and entangled xanthan in turbulent pipe flow, Journal of Non-Newtonian Fluid Mechanics, 166, 1-2: 25-31, doi.org/10.1016/j.jnnfm.2010.10.002. ##
[22]. Shams R, Shad S (2019) Experimental study of two-phase oil–polymer flow in horizontal flow path, Experimental Thermal and Fluid Science, 100: 62-75, doi.org/10.1016/j.expthermflusci.2018.08.028. ##
[23]. Nädler M, Mewes D (1997) Flow induced emulsification in the flow of two immiscible liquids in horizontal pipes, International Journal of Multiphase Flow, 23, 1: 55-68, doi.org/10.1016/S0301-9322(96)00055-9. ##
[24]. Li H, Wong T N, Skote M, Duan F (2014) Non-Newtonian two-phase stratified flow with curved interface through horizontal and inclined pipes, International Journal of Heat and Mass Transfer, 74: 113-120, doi.org/10.1016/j.ijheatmasstransfer.2014.02.052. ##
[25]. Langsholt M (2012) An experimental study on polymeric type DRA used in single-and multiphase flow with emphasis on degradation, diameter scaling and the effects in three-phase oil-water-gas flow, In 8th North American Conference on Multiphase Technology, OnePetro. ##
[26]. Lovick J, Angeli P (2004) Experimental studies on the dual continuous flow pattern in oil–water flows, International Journal of Multiphase Flow, 30, 2: 139-157, doi.org/10.1016/j.ijmultiphaseflow.2003.11.011. ##
[27]. Acharya T, Casimiro L (2020) Evaluation of flow characteristics in an onshore horizontal separator using computational fluid dynamics, Journal of Ocean Engineering and Science, 5, 3: 261-268, doi.org/10.1016/j.joes.2019.11.005. ##
[28]. Rabbani A, Babaei M, Shams R, Da Wang Y, Chung T (2020) DeePore: A deep learning workflow for rapid and comprehensive characterization of porous materials, Advances in Water Resources, 146: 103787, doi.org/10.1016/j.advwatres.2020.103787. ##
[29]. Goodfellow I, Bengio Y, Courville A (2016) Deep learning, MIT press. ##
[30]. Ershadnia R, Amooie M A, Shams R, Hajirezaie S, Liu Y, Jamshidi S, Soltanian M R (2020) Non-Newtonian fluid flow dynamics in rotating annular media: Physics-based and data-driven modeling, Journal of Petroleum Science and Engineering, 185: 106641, doi.org/10.1016/j.petrol.2019.106641. ##
[31]. Chang C W, Dinh N, Cetiner S M (2017) Physics-constrained machine learning for two-phase flow simulation using deep learning-based closure relation, In American Nuclear Society Winter Meeting, Washington, DC, 1749-1752. ##
[32]. Ezzatabadipour M, Singh P, Robinson M D, Guillén-Rondon P, Torres C (2017) Deep learning as a tool to predict flow patterns in two-phase flow, arXiv preprint arXiv:1705.07117, doi.org/10.48550/arXiv.1705.07117. ##
[33]. Raissi M, Yazdani A, Karniadakis G E (2018) Hidden fluid mechanics: A Navier-Stokes informed deep learning framework for assimilating flow visualization data, arXiv preprint arXiv:1808.04327. ##
[34]. Guillén-Rondon P, Robinson M D, Torres C, Pereya E (2018) Support Vector Machine Application for Multiphase Flow Pattern Prediction, arXiv preprint arXiv:1806.05054 https://doi.org/10.48550/arXiv.1806.05054. ##
[35]. Kanin E A, Osiptsov A A, Vainshtein A L, Burnaev E V (2019) A predictive model for steady-state multiphase pipe flow: Machine learning on lab data, Journal of Petroleum Science and Engineering, 180, 727-746, doi.org/10.1016/j.petrol.2019.05.055. ##
[36]. Gao Z, Hou L, Dang W, Wang X, Hong X, Yang X, Chen G (2020) Multitask-based temporal-channelwise CNN for parameter prediction of two-phase flows, IEEE Transactions on Industrial Informatics, 17, 9: 6329-6336. ##
[37]. Wang W, Gong J, Angeli P (2011) Investigation on heavy crude-water two phase flow and related flow characteristics, International Journal of Multiphase Flow, 37, 9: 1156-1164, doi.org/10.1016/j.ijmultiphaseflow.2011.05.011. ##
[38]. Oshinowo T, Charles M E (1974) Vertical two‐phase flow part I. Flow pattern correlations, The Canadian Journal of Chemical Engineering, 52, 1: 25-35, doi.org/10.1002/cjce.5450520105. ##
[39]. Mukherjee H, Brill J P (1983) Liquid holdup correlations for inclined two-phase flow, Journal of Petroleum Technology, 35, 05: 1003-1008, doi.org/10.2118/10923-PA. ##
[40]. Gao Z, Yang Y, Zhai L, Jin N, Chen G (2016) A four-sector conductance method for measuring and characterizing low-velocity oil–water two-phase flows, IEEE Transactions on Instrumentation and Measurement, 65, 7: 1690-1697, doi: 10.1109/TIM.2016.2540862. ##
[41]. LeCun Y (1989) Generalization and network design strategies, Connectionism in Perspective, 19, 143-155: 18. ##
[42]. LeCun Y, Bengio Y, Hinton G (2015) Deep learning, Nature, 521, 7553: 436-444. ##
[43]. Albion K, Briens L, Briens C, Berruti F (2008) Flow regime determination in horizontal hydrotransport using non‐intrusive acoustic probes, The Canadian Journal of Chemical Engineering, 86, 6: 989-1000, doi.org/10.1002/cjce.20112. ##
[44]. Tjugum S A, Hjertaker B T, Johansen G A (2002) Multiphase flow regime identification by multibeam gamma-ray densitometry, Measurement Science and Technology, 13, 8: 1319, doi:10.1088/0957-0233/13/8/321. ##
[45]. Otsu N (1979) A threshold selection method from gray-level histograms, IEEE Transactions on Systems, Man, and Cybernetics, 9, 1: 62-66. ##