پیوندهای مفید
اخبار و اعلانات
آمار
| تعداد نشریات | 127 |
| تعداد شمارهها | 7,196 |
| تعداد مقالات | 77,227 |
| تعداد مشاهده مقاله | 157,224,056 |
| تعداد دریافت فایل اصل مقاله | 118,406,883 |
بررسی عمق آبشستگی خطوط لوله عبوری از رودخانه با روشهای هوش مصنوعی | ||
| تحقیقات آب و خاک ایران | ||
| دوره 57، شماره 3، خرداد 1405، صفحه 567-587 اصل مقاله (1.43 M) | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22059/ijswr.2026.402770.670011 | ||
| نویسندگان | ||
| مهدی ماجدی اصل* 1؛ آرش فرجی1؛ توحید امیدپور علویان1؛ مهدی کوهدرق2 | ||
| 1University of Maragheh | ||
| 2گروه عمران، واحد ملکان. دانشگاه آزاد اسلامی، ملکان، ایران | ||
| چکیده | ||
| آبشستگی موضعی اطراف خطوط لوله یکی از عوامل مهم ناپایداری بستر و آسیب به زیرساختهای انتقال سیال است که میتواند پیامدهای فنی و زیستمحیطی جدی به همراه داشته باشد. در این پژوهش، کارایی سه روش هوش مصنوعی شامل شبکه عصبی مصنوعی، ماشین بردار پشتیبان و کیونت در پیشبینی عمق آبشستگی با استفاده از ۳۶ مجموعه داده آزمایشگاهی ارزیابی شد. پارامترهای ورودی شامل قطر لوله، فاصله از بستر، عمق جریان، عدد فرود و طول لوله بوده و دقت مدلها با شاخصهای آماری مختلف بررسی گردید. نتایج هوش مصنوعی نشان داد که روش شبکه عصبی مصنوعی (روش برتر) در مرحله آموزش مربوط به ترکیب برتر با شاخصههای مورد ارزیابی ریشه میانگین مربعات خطا، مجذور ضریب همبستگی و ضریب تبیین در مراحل آموزش و آزمون به ترتیب برابر است با 0272/0، 9932/0، 9925/0 و 1180/0، 8959/0، 8935/0 میباشد. نتایج آنالیز حساسیت نیز نشان داد که حذف پارامتر فاصله لوله از بستر منجر به کاهش دقت مدل و افزایش میزان خطا شده است. این نتیجه بیانگر این است که تغییرات فاصله لوله از بستر به طور مستقیم بر عمق آبشستگی تأثیر میگذارد. این مطالعه نشان میدهد که هر سه روش هوش مصنوعی با دقت بالا میتوانند جایگزین مناسبی برای مدلهای تجربی سنتی باشند و میتوانند در کاهش ریسکها و هزینههای مرتبط با خرابی سازههای زیرآبی و مدیریت بهینه زیرساختهای هیدرولیکی مؤثر باشند. | ||
| کلیدواژهها | ||
| شبکههای عصبی مصنوعی (ANN)؛ فرسایش؛ مدل QNET؛ پارامترهای آماری؛ ماشین بردار پشتیبان (SVM) | ||
| عنوان مقاله [English] | ||
| Investigation of Scour Depth Around Pipelines Crossing Rivers Using Artificial Intelligence Methods | ||
| نویسندگان [English] | ||
| Mahdi Majedi Asl1؛ Arash faraji1؛ tohid omidpour alavian1؛ mehdi Kouhdaragh2 | ||
| 1University of Maragheh | ||
| 2Civil Engineering Department, Engineering Faculty, Malekan Branch, Islamic Azad University, Malekan, Iran | ||
| چکیده [English] | ||
| Local scour around pipelines is one of the main causes of bed instability and damage to fluid conveyance infrastructures, which can lead to serious technical and environmental consequences. In this study, the performance of three artificial intelligence methods, including Artificial Neural Network (ANN), Support Vector Machine (SVM), and QNET, was evaluated for predicting scour depth using 36 experimental datasets. Input parameters included pipe diameter, pipe embedment depth, flow depth, Froude number, and pipe length. The accuracy of the models was assessed using various statistical indices. Model performance was assessed using the coefficient of determination (R²), root mean square error (RMSE), and the Nash–Sutcliffe coefficient (DC). Results indicated that the Artificial Neural Network (ANN) outperformed the other models, achieving the following optimal values for the best architecture in the training and testing phases, respectively: RMSE = 0.0272, R² = 0.9932, DC = 0.9925 and RMSE = 0.1180, R² = 0.8959, DC = 0.8935. Sensitivity analysis further demonstrated that removing the pipeline-to-bed gap parameter significantly reduced model accuracy and increased prediction error, highlighting the dominant direct influence of this parameter on scour depth. The study concludes that all three AI-based approaches provide high predictive accuracy and can effectively replace conventional empirical models, thereby contributing to risk reduction, cost savings in submarine and river-crossing pipeline projects, and improved management of hydraulic infrastructure. | ||
| کلیدواژهها [English] | ||
| Artificial Neural Networks (ANN), Errosion, QNET Model, Statistical Parameters, Support Vector Machines (SVM) | ||
| مراجع | ||
|
Abbasi, M., & Kazemi, F. (2018). Limitations and adjustments of QNET in complex flow scenarios. Applied Ocean Research, 75, 76–82. Atayeian, A. (2011). Estimation of local scour depth beneath a pipeline crossing a river [Master’s thesis, University of Urmia, Faculty of Agriculture, Iran]. (In persian) Azamatullah, M. N., Deo, M. C., & Yadav, R. N. (2012). Prediction of pipeline scour using neural networks. Marine Structures, 25(1), 34–53. Bateni, S. M., Borghei, S. M., & Jeng, D. S. (2007). Neural network and neuro-fuzzy assessments for scour depth around bridge piers. Advances in Engineering Software, 38(2), 102–111. https://doi.org/10.1016/j.engappai.2006.06.012 Bijker, E. W., & Leeuwestein, W. (1984). Interaction between pipelines and the seabed under the influence of waves and currents. Proceedings of the Conference on Behavior of Offshore Structures, 3, 343–356. https://doi.org/10.1007/978-94-009-4958-4_28 Chao, J. L., & Hennessy, P. V. (1972). Local Scour under Ocean Outfall Pipelines. Journal (Water Pollution Control Federation), 44(7), 1443–1447. http://www.jstor.org/stable/25037552 Chiew, Y. M. (1991). Prediction of maximum scour depth at submarine pipelines. Journal of Hydraulic Engineering, 117(4), 452–466. https://doi.org/10.1061/(ASCE)0733429(1991)117:4(452) Choi, S. U., & Jung, S. (2006). Prediction of local scour around bridge piers using artificial neural networks. Journal of the American Water Resources Association, 42(2), 487–494. https://doi.org/10.1111/j.1752-1688.2006.tb03852.x De, D., Roy, S., & Chattopadhyay, S. (2006). Predicting scour depth using neural networks. Journal of Hydrologic Engineering, 11(2). Dey, X. (2010). Comparative analysis of scour prediction methods: The QNET approach. International Journal of Hydraulic Research, 52(1), 120–130. Dey, X., Zhao, Y., & Lu, Q. (2008). Mechanics of steady-state scour: a comparative study using QNET models. Journal of Hydraulic Engineering, 134(11), 1608–1616. Ferdesou, T. (2020). QNET model enhancement for pipeline-riverbed interaction analysis. Journal of Marine Science and Engineering, 8(11), 890. Habib, M. A., Abolfathi, S., O’Sullivan, John. J., & Salauddin, M. (2024). Efficient data-driven machine learning models for scour depth predictions at sloping sea defences. Frontiers in Built Environment, 10, 1343398. https://doi.org/10.3389/fbuil.2024.1343398 . (In persian) Hassan, W. H., & Jalal, H. K. (2021). Prediction of the depth of local scouring at a bridge pier using a gene expression programming method. SN Applied Sciences, 3(2), 159. https://doi.org/10.1007/s42452-020-04124-9. (In persian) Haykin, S. (1999). Neural Networks: A Comprehensive Foundation. Prentice Hall. Heydari, S., et al. (2016). Scour prediction under storm conditions using QNET: A tool for natural disaster management. Natural Hazards Review, 17(2), 04016006. Hu, K., Bai, X., Zhang, Z., & Vaz, M. A. (2021). Prediction of submarine pipeline equilibrium scour depth based on machine learning applications considering the flow incident angle. Applied Ocean Research, 112, 102717. https://doi.org/10.1016/j.apor.2021.102717 Hu, R., Wang, X., Liu, H., Leng, H., & Lu, Y. (2022). Scour Characteristics and Equilibrium Scour Depth Prediction around a Submarine Piggyback Pipeline. Journal of Marine Science and Engineering, 10(3), 350. https://doi.org/10.3390/jmse10030350 Ibrahim, A., & Nalluri, C. (1986). Scour prediction around marine pipelines. In International offshore mechanics and arctic engineering. Symposium. 5 (pp. 679–684). Kazilos, P., et al. (2011). Utilization of QNET for enhanced scour prediction around submarine pipelines. Marine Structures, 24(3), 318–329. Khan, Z., Nanda, S., & Goel, A. (2014). Prediction of scour depth around submerged pipelines using artificial neural networks. Journal of Pipeline Systems Engineering and Practice, 5(1), 04013009. Kjeldsen, S. P., Einstein, H. H., & Christensen, B. A. (1973). Local scour near offshore pipelines. In Paper available only as part of the complete Proceedings of the Second International Conference on Port and Ocean Engineering Under Arctic Conditions (POAC), August 27-30, 1973. Marulasiddappa, S. B., Patil, A. P., Kuntoji, G., et al. (2024). Prediction of scour depth around bridge abutments using ensemble machine learning models. Neural Computing and Applications, 36, 1369–1380. https://doi.org/10.1007/s00521-023-09109-4 Mattioli, P., et al. (2019). Bed dynamics and QNET modeling accuracy in close-bed interactions. Journal of Waterway, Port, Coastal, and Ocean Engineering, 145(5), 04019012. Maza, J. A. (1987). Introduction to river engineering. Advanced Course on Water Resources Management, Universita Italiana per Stranieri. Nandi, B., & Das, S. (2025). Prediction of maximum scour depth at clear water conditions: Multivariate analysis. Journal of Environmental Management, 354, 120349. https://doi.org/10.1016/j.jenvman.2024.120349 Oliveira, A., Melo, M. S., & Tormena, J. M. (2013). Application of artificial neural networks for scour prediction at submerged pipeline crossings. Applied Ocean Research, 40, 75–82. Parsaie, A., Haghiabi, A. H., & Moradinejad, A. (2019). Prediction of Scour Depth below River Pipeline using Support Vector Machine. KSCE Journal of Civil Engineering, ۲۳(۶), ۲۵۶۲–۲۵۷۱. https://doi.org/10.1007/s12205-019-1327-0 Qnet v2000 (Trial Version). (2000). Developed by Vesta Services, Inc., Winnetka, IL, USA. Archival information and download links available at: https://qnet-2000.software.informer.com/ and https://qnetv2kt.software.informer.com Rahimi, A., et al. (2015). Application of QNET for turbulent flow conditions: Environmental and structural impact analysis. Journal of Hydraulic Engineering, 141(8), 04015019. Rahman, F., & Chavan, R. (2025). Machine Learning Application in Prediction of Scour Around Bridge Piers: A Comprehensive Review. Archive of Computational Methods in Engineering, 32, 1299–1322. https://doi.org/10.1007/s11831-024-10167-7 Sharifi, M., et al. (2017). Evaluating QNET for scour depth prediction across varied flow and substrate conditions. Journal of Coastal Research, 34(1), 1–10. Soltani Sotobadi, M., Omidpour Alavian, T., & Kardan, N. (2025). Forecasting the Local Scour Depth around Cylindrical Bridge Foundations on Cohesive Soil Utilizing Meta Models. Iranian Journal of Irrigation and Drainage, 19(2), 275–290. (In persian) Vapnik, V. N. (1995). The Nature of Statistical Learning Theory. Springer-Verlag. https://doi.org/10.1007/978-1-4757-2440-0 Xiu, Z., Luo, C., Liu, L., Du, X., Gao, W., Song, Y., Shi, B., & Chi, W. (2025). CFD and machine learning approach based-predictive modeling of scouring below submarine pipeline under wave and current condition. Marine Georesources & Geotechnology, 43(9), 1779-1795. https://doi.org/10.1080/1064119X.2024.2434954 | ||
|
آمار تعداد مشاهده مقاله: 725 تعداد دریافت فایل اصل مقاله: 88 |
||
سامانه مدیریت نشریات علمی. قدرت گرفته از سیناوب