نوع مقاله : مقاله پژوهشی

نویسنده

استادیار، گروه عمران، دانشکده مهندسی، دانشگاه آزاد اسلامی، واحد بوکان، بوکان، ایران

چکیده

در این پژوهش از دو مدل یادگیری ماشین شامل LS-SVR و ANFIS برای پیش­بینی عمق آب شستگی اطراف پایه­های پل استفاده شد. برای این منظور از 240 سری داده شامل پارامترهای مرتبط با هندسه پایه ها، شرایط جریان و خصوصیات جریان و نیز پارامترهای بدون بعد استفاده شد. برای پیش­بینی از دو الگوی ورودی استفاده شد. در الگوی اول، پارامترهای بدون بعد و در الگوی دوم پارامترهای با بعد در نظر گرفته شدند. عملکرد مدل‌ها با استفاده از معیارهای ریشه میانگین مربعات خطا (RMSE)، میانگین درصد مطلق خطا (MAPE) و ضریب نش-ساتکلیف (NSE) ارزیابی شدند. نتایج نشان داد که در هر دو مدل، استفاده از پارامترهای با بعد برای پیش­بینی منجر به‌دقت بالای پیش­بینی می‌شود. مقایسه بین مدل‌ها نیز نشان داد که الگوریتم LS-SVR  با معیارهایRMSE=46.84, MAPE=38.03 , NSE=0.62 برای داده­های آزمون الگوی اول و RMSE=28.62 , MAPE=38.97 , NSE=0.67 برای داده­های آزمون الگوی دوم دقت بالاتری نسبت به الگوریتم ANFIS دارد. نتایج این تحقیق حاکی از این است که مدل­های یادگیری ماشین جایگزین مناسبی برای مدل­های تجربی در پیش­بینی عمق آبشستگی پایه­های پل هستند.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Performance Evaluation of LS-SVR Model in Predicting Scour Depth in Bridge Piers

نویسنده [English]

  • Bijan Sanaati

Assist. Professor, Department of Civil Engineering, Faculty of Engineering, Islamic Azad University, Bukan Branch, Bukan, Iran

چکیده [English]

In this research work, two machine learning models including Least Squares Support Vector Machines (LS-SVR) and Adaptive Neuro-Fuzzy Inference System (ANFIS) were used to predict the scour depth around the bridge piers. For this purpose, 240 data series including pier geometry, flow condition, sediment characteristics, and some dimensional parameters were used. Dimensional and no dimensional parameters were considered. The performance of the models was evaluated using root mean square error (RMSE), mean absolute percentage error (MAPE), and Nash–Sutcliffe efficiency (NSE) criteria. The results showed that in both models, the use of dimensional parameters for prediction leads to high prediction accuracy. The comparison between the models also showed that the LS-SVR algorithm with the criteria RMSE=46.84, MAPE=38.03, NSE=0.62 for the test data of the first model and RMSE=28.62, MAPE=38.97, NSE=0.67 for the test data results of the second pattern are more accurate than the ANFIS algorithm. This research indicates that machine learning models are a suitable alternative to empirical models in predicting scour depth of bridge piers.

کلیدواژه‌ها [English]

  • ANFIS
  • Bridge Piers
  • LS-SVR
  • Machine Learning
  • Prediction
  • Scour Depth
Azar, N. A., Milan, S. G., & Kayhomayoon, Z. (2021). The prediction of longitudinal dispersion coefficient in natural streams using LS-SVM and ANFIS optimized by Harris hawk optimization algorithm. J. Contamin. Hydrol., 240, 103781.  DOI: 10.1016/j.jconhyd.2021.103781.
Azamathulla, H. M. (2012). Gene expression programming for prediction of scour depth downstream of sills. J. Hydrol., 460, 156–159.  DOI: 10.1016/j.jhydrol.2012.06.034.
Bateni, S. M., Borghei, S. M., & Jeng, D. S. (2007). Neural Network and neurofuzzy assessments for scour depth around bridge piers. Eng. Appl. Artif. Intell., 20(3), 401-414.  DOI: 10.1016/j.engappai.2006.06.012.
Chou, J. S., & Nguyen, N. M. (2022). Scour depth prediction at bridge piers using metaheuristics-optimized stacking system. Autom. Construct., 140, 104297.  DOI: 10.1016/j.autcon.2022.104297.
Dargahi, B. (1990). Controlling mechanism of local scour. J. Hydraul. Eng., 116, 1197–1214. DOI: 10.1061/(ASCE)0733-9429(1990).
Dodaro, G., Tafarojnoruz, A., Calomino, F., Gaudio, R., Stefanucci, F., Adduce, C., & Sciortino, G.  (2014). An experimental and numerical study on the spatial and temporal evolution of a scour hole downstream of a rigid bed. In: Proceedings of the International Conference on Fluvial Hydraulics, RIVER FLOW 2014. Taylor and Francis Group plc, Lausanne, Switzerland, pp. 1415–1422. DOI: 10.1201/b17133-189.
Dodaro, G., Tafarojnoruz, A., Sciortino, G., Adduce, C., Calomino, F., & Roberto, G. (2016).  Modified Einstein sediment transport method to simulate the local scour evolution downstream of a rigid bed. J. Hydraul. Eng., 142, 4016041. DOI: 10.1061/(asce)hy.1943-7900.0001179.
Ettema, R., Melville, B. W., & Barkdoll, B. (1998).  Scale effect of pier-scour experiments. J. Hydraul. Eng., 124, 639– 642. DOI: 10.1061/(asce)0733-9429(1998)124:6(639).
Ettema, R., Constantinescu, G., & Melville, B. W. (2017). Flow-field complexity and design estimation of pier-scour depth: Sixty years since Laursen and Toch. J. Hydraul. Eng., 143(9), 03117006. DOI: 10.1061/(asce)hy.1943-7900.0001330.
Ghafari, H., & Zomorodian, M. A. (2019). Investigating the local scour around group bridge piers in cohesive Soils. Journal of Water and Soil Science, 23(4), 109-123. [In Persian]
Ghordoyee Milan, S., Aryaazar, N., Javadi, S., & Razdar, B. (2020). Simulation of groundwater head using LS-SVM and comparison with ANN and MLR. Hydrogeol., 5(1), 118-133. [In Persian]
Jafari Bavil Olyaei, A., hassanzadeh, Y., Alami, M., and kardan, N. (2018). Estimation of Bridge Pier Scour using Adaptive Neuro-Fuzzy Inference System Optimized with Imperialist competitive algorithm. Iran. J. Irrig. Drain., 12(4), 872-884.
Jang, J. S., (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE Trans.
Syst. Man. Cybern., 23(3),
665–685. DOI: 10.1109/21.256541.
Johnson, P.  A., (1992). Reliability-based pier scour engineering. J. Hydraul. Eng., 118, 1344–1358. DOI: 10.1061/(ASCE)0733-9429(1992)118:10(1344).
Karami Moghadam, M., & Sabzevari, T. (2018). Modification of Bridge Piers Scour Depth Equations using Genetic Algorithm. Environ. Water Eng., 4(2), 101-114. DOI: 10.22034/jewe.2018.112907.1223.
Kayhomayoon, Z., Naghizadeh, F., Malekpoor, M., Arya Azar, N., Ball, J., & Ghordoyee Milan, S. (2022). Prediction of evaporation from dam reservoirs under climate change using soft computing techniques. Environ. Sci. Pollut. Res., 30(10), 27912. DOI: 10.1007/s11356-022-23899-5.
Kirkil, G., Constantinescu, S. G., & Ettema, R. (2008). Coherent structures in the flow field around a circular cylinder with scour hole. J. Hydraul. Eng., 134, 572–587. DOI: 10.1061/(asce)0733-9429(2008)134:5(572).
Koopaei, K. B., & Valentine, E.M. (2003). Bridge Pier Scour in Self Formed Laboratory Channels; Technical Report; University of Glasgow: Glasgow, UK.
Kuhn, H. W., & Tucker, A. W. (1951). Nonlinear programming. Proceedings of 2nd Berkeley Symposium. Berkeley: University of California Press. pp. 481–492. MR 0047303.
Majedi Asl, M., & Valizadeh, S. (2019). Application of SVM algorithm in predicting vertical pier scour depth. J. Water Soil Sci., 23(4), 165-181. [In Persian]
Mobayen, R., Najafzadeh, M., & Farrahi-Moghaddam, K. (2023). Evaluation of regression-based soft computing techniques for estimating energy loss in gabion spillways. Environ. Water Eng., 9(2), 241-255. [In Persian]. DOI: 10.22034/ewe.2022.329153.1724.
Najafzadeh, M., & Azamathulla, H. M. (2015). Neuro-fuzzy GMDH to predict the scour pile groups due to waves. J. Comput. Civil Eng., 29(5), 04014068. DOI: 10.1061/(asce)cp.1943-5487.0000376.
Richardson, E., & Davis, S. (2001). Evaluating Scour at Bridges: Hydraulic Engineering Circular. FHWA-IP-90-017, HEC-18. DOI: 10.1061/41147(392)110.
Samadi, M., Afshar, M. H., Jabbari, E., & Sarkardeh, H. (2021). Prediction of current-induced scour depth around pile groups using MARS, CART, and ANN approaches. Mar. Georesour. Geotechnol., 39(5), 577-588. DOI: 10.1080/1064119x.2020.1731025.
Shamshirband, S., Mosavi, A., & Rabczuk, T. (2020). Particle swarm optimization model to predict scour depth around a bridge pier. Front. Struct. Civil Eng., 14, 855-866. DOI: 10.1007/s11709-020-0619-2.
Sreedhara, B. M., Rao, M., & Mandal, S. (2019). Application of an evolutionary technique (PSO–SVM) and ANFIS in clear-water scour depth prediction around bridge piers. Neur. Comput. Appl., 31, 7335-7349.DOI: 10.1007/s00521-018-3570-6.
Suykens, J. A., & Vandewalle, J. (1999). Least squares support vector machine classifiers.
Neur. Process. Lett., 9(3), 293–300. DOI: 10.1109/ijcnn.1999.831072.