پیش بینی میزان ماسه عبوری از لوله ی مشبک کاری شده به کمک الگوریتم های یادگیری ماشین

خداشناس, محمدالیاس

doi:20.1001.1/ijge.2026.2088830.1125

پیش بینی میزان ماسه عبوری از لوله ی مشبک کاری شده به کمک الگوریتم های یادگیری ماشین

نوع مقاله : مقاله پژوهشی

نویسنده

محمدالیاس خداشناس

دانشجوی کارشناسی ارشد، مهندسی نفت، دانشکده مهندسی شیمی، نفت و گاز، دانشگاه شیراز، شیراز، ایران

20.1001.1/ijge.2026.2088830.1125

چکیده

تولید ماسه از چاه‌های نفت و گاز باعث کاهش بهره‌وری و خرابی تجهیزات تولیدی می‌شود. برای کنترل این پدیده از توری‌های مهارکننده مکانیکی به‌ویژه لوله مشبک‌کاری شده استفاده می‌شود که عملکرد این توری‌ به‌وسیله‌ی تحلیل نتایج آزمون نگهداشت ماسه تعیین می‌شود. در این پژوهش به‌منظور پیش‌بینی میزان ماسه عبوری از واحد سطح توری لوله مشبک‌کاری شده، از الگوریتم‌های یادگیری ماشین استفاده ‌شده است. بدین منظور، نتایج حاصل از ۵۱۹ نمونه آزمایش آزمون نگهداشت ماسه‌ی ذکرشده در مطالعات پیشین به‌عنوان پایگاه داده جمع‌آوری شد که شامل: انواع ساختار و طراحی‌های مختلف توری، توزیع اندازه ذرات ماسه و شرایط آزمایشی است. پایگاه داده‌ی نهایی شامل: ۴۹۹ نمونه داده‌ی پیش‌پردازش شده است. برای ساخت مدل از الگوریتم‌های: GB، XGBoost، RF، DT، KNN، SVM، MLP و Poly Reg استفاده شد. نتایج نشان داد که الگوریتم‌ GB با بیشترین دقت ۹۴/۰R2=) و کمترین خطا، به‌همراه الگوریتم XGBoost بهترین عملکرد را در پیش‌بینی ارائه می‌دهند. همچنین مشخص شد که اندازه عرض شکاف‌های توری و اختلاف فشار و دبی آزمایش مهم‌ترین عوامل تأثیرگذار بر مقدار ماسه تولیدی می-باشند. به‌علاوه، نوع طراحی توری و اندازه عرض شکاف‌های آن باید متناسب با توزیع اندازه ذرات ماسه انتخاب شود. مدل ماشینی این پژوهش، مزایایی همچون: افزایش دقت پیش‌بینی نسبت به روش‌های پیشین، ارائه‌ی یک مدل تفسیرپذیر عملکرد توری و افزایش سرعت انتخاب را به همراه دارد و می‌تواند به کنترل تولید ماسه‌ی چاه‌ها کمک کند.

کلیدواژه‌ها

آزمون نگهداشت ماسه

تکمیل چاه

تولید ماسه

لوله مشبک‌کاری شده

یادگیری ماشین

موضوعات

مهندسی نفت

عنوان مقاله English

Prediction of Sand Passage Rate Through Slotted Liner Using Machine Learning Algorithms

نویسنده English

Mohammadelyas Khodashenas

M.Sc. Student, Petroleum Engineering, School of Chemical Engineering, Oil and Gas, Shiraz University, Shiraz, Iran

چکیده English

Sand production from oil and gas wells reduces productivity and causes damage to production equipment. To control this phenomenon, mechanical sand screens, particularly slotted liners, are employed, and the performance of these screens is determined through the analysis of sand retention test results. In this study, machine learning algorithms were utilized to predict the amount of sand passing through the unit surface area of slotted liner coupons. To this end, the results of 519 sand retention test samples reported in previous studies were compiled as a dataset, encompassing various screen structures and designs, sand particle size distributions, and experimental conditions. The final dataset consisted of 499 preprocessed data samples. The models were constructed using the following algorithms: GB, XGBoost, RF, DT, KNN, SVM, MLP, and Polynomial Regression. The results demonstrated that the GB algorithm, with the highest accuracy (R² = 0.94) and the lowest error, alongside the XGBoost algorithm, delivered the best predictive performance. Furthermore, it was found that slot width, pressure differential, and flow rate are the most influential factors affecting the amount of produced sand. Additionally, the screen design type and slot width must be selected in accordance with the particle size distribution of the sand. The machine learning model developed in this research offers advantages such as enhanced prediction accuracy compared to previous methods, providing an interpretable model of screen performance, and increasing the speed of selection, thereby contributing to the control of sand production in wells.

کلیدواژه‌ها English

Sand retention test

Well completion

Sand production

Slotted liner

Machine learning

[1] Zare-Reisabadi M, Safari-Beidokhti M, Kaffash A, Bataee M. Sand production prediction and well completion optimization. International Journal of Petroleum and Geoscience Engineering (IJPGE). 2014;2(4):361-74.
[2] Kumar S, Kumar G, Krishna S, Kumar A, Gupta T, Goel PN, Kumari S. Sand screens for controlling sand production from hydrocarbon wells: A mini-review. Geoenergy Science and Engineering. 2024;240:213040.
[3] Velayati A, Roostaei M, Fattahpour V, Mahmoudi M, Nouri A, Alkouh A, et al., editors. Design Optimization of Slotted Liner Completions in Cased and Perforated Wells: A Numerical Skin Model. SPE International Heavy Oil Conference and Exhibition; 2018: SPE.
[4] Wang C, Pang Y, Mahmoudi M, Haftani M, Salimi M, Fattahpour V, Nouri A. A set of graphical design criteria for slotted liners in steam assisted gravity drainage production wells. Journal of Petroleum Science and Engineering. 2020;185:106608.
[5] Guo Y, Nouri A, Nejadi S. Effect of slot width and density on slotted liner performance in SAGD operations. Energies. 2020;13(1):268.
[6] Roostaei M, Cespedes EAM, Uzcátegui AA, Soroush M, Hosseini SA, Izadi H, et al. Optimization of slotted liner in rubiales field: trade-off between sand control, flow performance, and plugging tendency. SPE Journal. 2021; 26(03)30-1110.
[7] Kaiser T, Wilson S, Venning L. Inflow analysis and optimization of slotted liners. SPE Drilling & Completion. 2002;17(04):200-9.
[8] Abduljabbar A, Amadi A, Mohyaldinn ME, Ridha S, Younis O, Alakbari FS. Sand screens application and performance for sand control: A review of selection criteria, screen materials, and causes of failure. Heliyon. 2024;10 (10).
[9] Haftani M, Kotb O, Nguyen PH, Wang C, Salimi M, Nouri A. A Novel sand control testing facility to evaluate the impact of radial flow regime on screen performance and its verification. Journal of Petroleum Science and Engineering. 2020;195:107903.
[10] Wu B, Choi S, Feng Y, Denke R, Barton T, Wong C, et al., editors. Evaluating sand screen performance using improved sand retention test and numerical modelling. Offshore technology conference Asia; 2016: OTC.
[11] Reyes R, Sipi A, editors. Optimal sand control design & technique selection: a simplified practical guidance tool. Offshore Technology Conference Asia; 2018: OTC.
[12] Ballard T, Beare S, editors. An investigation of sand retention testing with a view to developing better guidelines for screen selection. SPE International Conference and Exhibition on Formation Damage Control; 2012: SPE.
[13] Ballard T, Beare SP, editors. Sand retention testing: the more you do, the worse it gets. SPE International Conference and Exhibition on Formation Damage Control; 2006: SPE.
[14] Hassani H, Silva ES. Big Data: a big opportunity for the petroleum and petrochemical industry. OPEC Energy Review. 2018;42(1):74-89.
[15] Chaudhary S, Yadav S, Kushwaha S, Shahi SRP. A brief review of machine learning and its applications. SAMRIDDHI: A Journal of Physical Sciences, Engineering and Technology. 2020;12(SUP 1):218-23.
[16] Ray S, editor A quick review of machine learning algorithms. 2019 International conference on machine learning, big data, cloud and parallel computing (COMITCon); 2019: IEEE.
[17] Razak NNA, Abdulkadir SJ, Maoinser MA, Shaffee SNA, Ragab MG. One-dimensional convolutional neural network with adaptive moment estimation for modelling of the sand retention test. Applied Sciences. 2021;11(9):3802.
[18] Ngwashi AR, Ogbe DO, Udebhulu DO, editors. Evaluation of machine-learning tools for predicting sand production. SPE Nigeria Annual International Conference and Exhibition; 2021: SPE.
[19] Asfha DT, Latiff AHA, Otchere DA, Tackie-Otoo BN, Babikir I, Rafi M, et al. Mechanisms of sand production, prediction–a review and the potential for fiber optic technology and machine learning in monitoring. Journal of Petroleum Exploration and Production Technology. 2024;14(10):2577-616.
[20] Azad M, Zargar G, Arabjamaloei R, Hamzei A, Ekramzadeh M. A new approach to sand production onset prediction using artificial neural networks. Petroleum science and technology. 2011;29(19):1975-83.
[21] Hamid ZHA, Hernandez N, Baillard F, Looi T, editors. Sand Production and Control Benchmarking through Unstructured Data Analysis with Machine Learning in the North Sea. 84th EAGE Annual Conference & Exhibition; 2023: European Association of Geoscientists & Engineers.
[22] Laoufi H, Megherbi Z, Zeraibi N, Merzoug A, Ladmia A, editors. Selection of sand control completion techniques using machine learning. ARMA/DGS/SEG International Geomechanics Symposium; 2022: ARMA.
[23] Song J, Li Y, Liu S, Xiong Y, Pang W, He Y, Mu Y. Comparison of machine learning algorithms for sand production prediction: an example for a gas-hydrate-bearing sand case. Energies. 2022;15(18):6509.
[24] Laoufi H, Megherbi Z, Zeraibi N, Merzoug A, Ladmia A, editors. Selection of Sand Control Completion Techniques Using Machine Learning. International Geomechanics Symposium; 2022. ARMA-IGS-2022-118.
[25] Khodashenas M, Mohammadzadeh Shirazi M, Shahsavani B. Application of machine learning in sand production prediction and management in oil and gas wells: a review. Iranian Gas Engineering Journal. 2025;12(1):54-69. [In Persian]
[26] Toro C, Pietrangeli G, Lara J, Pineda M, Mesa S, editors. Innovative Remediation System for Mature-Heavy Crude Oil Fields Completed with Slotted Liners and Screens. Offshore Technology Conference Brasil; 2025: OTC.
[27] Xie J, editor Slotted liner design optimization for sand control in SAGD wells. SPE Thermal Integrity and Design Symposium; 2015: SPE.
[28] Mahmoudi M, Fattahpour V, Wang C, Kotb O, Roostaei M, Nouri A, et al., editors. A Large-Scale Sand Retention Test Facility for Evaluation and Selection of Optimal Standalone Sand Screen for Injection Wells in Thermal Operations. SPE International Conference and Exhibition on Formation Damage Control; 2018: SPE.
[29] Mayorga Cespedes EA, Roostaei M, Uzcátegui AA, Soroush M, Izadi H, Hosseini SA, et al., editors. Sand control optimization for rubiales field: trade-off between sand control, flow performance and mechanical integrity. SPE Latin America and Caribbean Petroleum Engineering Conference; 2020: SPE.
[30] Ochmann L, Joseph R. Development of a sand retention test set-up focusing on the measurement of produced sand. OIL GAS European Magazine. 2021;2:25-32.
[31] Fattahpour V, Roostaei M, Hosseini SA, Soroush M, Berner K, Mahmoudi M, et al. Experiments with stand-alone sand-screen specimens for thermal projects. SPE Drilling & Completion. 2021;36(01):188-207.
[32] Mahmoudi M, Fattahpour V, Nouri A, Yao T, Baudet BA, Leitch M, Fermaniuk B, editors. New criteria for slotted liner design for heavy oil thermal production. SPE Thermal Integrity and Design Symposium; 2016: SPE.
[33] Mahmoudi M, Fattahpour V, Nouri A, Leitch M, editors. An experimental evaluation of pore plugging and permeability reduction near SAGD sand control liners. SPE Canada Heavy Oil Conference; 2017: SPE.
[34] Devere-Bennett N, editor Using prepack sand-retention tests (SRT's) to narrow down liner/screen sizing in SAGD wells. SPE Thermal Integrity and Design Symposium; 2015: SPE.
[35] Fischer C, Hamby H, editors. A novel approach to constant flow-rate Sand retention testing. SPE International Conference and Exhibition on Formation Damage Control; 2018: SPE.
[36] Ismail NI, Kuang S, Zheng E, Yu A. Modeling and analysis of fluid rheology effect on sand screen performance. Powder Technology. 2022;411:117961.
[37] Cabello-Solorzano K, Ortigosa de Araujo I, Peña M, Correia L, J. Tallón-Ballesteros A, editors. The impact of data normalization on the accuracy of machine learning algorithms: A comparative analysis. International conference on soft computing models in industrial and environmental applications; 2023: Springer.
[38] Chicco D, Warrens MJ, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. Peerj computer science. 2021;7:e623.
[39] Burzykowski T, Geubbelmans M, Rousseau A-J, Valkenborg D. Validation of machine learning algorithms. American journal of orthodontics and dentofacial orthopedics.
[40] Uyeh DD, Iyiola O, Mallipeddi R, Asem-Hiablie S, Amaizu M, Ha Y, Park T. Grid search for lowest root mean squared error in predicting optimal sensor location in protected cultivation systems. Frontiers in Plant Science. 2022;13:9202.