بررسی کارآیی الگوریتم M5 در محاسبه حداکثر عمق چاله آبشستگی اطراف تکیهگاه پل

نوع مقاله: مقاله پژوهشی

نویسندگان

1 استادیار گروه مهندسی آب دانشگاه علوم کشاورزی و منابع طبیعی رامین خوزستان

2 استاد دانشگاه شهید چمران اهواز

چکیده

آبشستگی صورت گرفته در اطراف تکیهگاهها و پایهها پدیدهای است که منجر به تخریب تعداد زیادی از پلها گردیده است. در این تحقیق برای شبیهسازی حداکثر عمق چاله آبشستگی اطراف تکیه­گاه پل از الگوریتم درختی M5 استفاده شده است. از مزایای این الگوریتم دقت بالا، سادگی و قابل فهم بودن آن است. برای ساخت و آموزش مدل از اطلاعات آزمایشگاهی منابع معتبر استفاده شد. پارامترهای مورد استفاده برای برآورد عمق آبشستگی اطراف تکیهگاه شامل فاکتور شکل و طول تکیهگاه، عمق و سرعت جریان و اندازه متوسط رسوبات بستر بوده است که به­صورت عدد فرود ذره، ضریب شکل تکیهگاه، ضریب عمق طول تکیهگاه و ضریب شدت جریان به مدل درختی معرفی شدند. در مدل ساخته شده توسط الگوریتم M5 و پس از عملیات هرس کردن، ضریب عمق طول تکیهگاه و ضریب شدت جریان بیشترین کاربرد را در ساختار درختی داشته و باعث تقسیم فضای مسأله به سه زیردامنه متمایز و ارائه رابطه رگرسیونی برای هر زیردامنه گشته است. برای بررسی کارآیی مدل درختی از تحلیل­های آماری متفاوت و مقایسه آن با معادلات ارائه شده توسط محققین مختلف استفاده شد. نتیجه این مقایسه نشاندهنده کارآیی بالای الگوریتم M5 در محاسبه عمق آبشستگی میباشد. برای بررسی اهمیت هر کدام از پارامترهای ورودی بر میزان آبشستگی از تحلیل حساسیت استفاده گردید. بر این اساس ضریب عمق طول تکیهگاه بیشترین تأثیر را بر میزان آبشستگی داشته و حذف این پارامتر از مدل درختی باعث افزایش مجموع مربعات خطا از 37/4 به 41/19 گشته است.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Predicting maximum scour depth around bridge abutment using M5 model

نویسندگان [English]

  • Javad Zahiri 1
  • Seyed Mahmood Kashefipour 2
1 Assistant Professor, Department of Water Engineering, Khuzestan Ramin Agriculture and Natural Resources University, Iran
2 Professor, Water Sciences Engineering Faculty, Shahid Chamran University, Ahwaz, Iran
چکیده [English]

Scour around piers and abutments leads to a lot of bridge failures. Scouring around the bridge piers and abutments is one of the most important factors in bridge failure which happens over time and increases during floods, probably causing general damage to the bridge. Accordingly, the accurate estimation of scouring around the bridge piers and abutments can help engineers to take steps to deal with bridge destructions. Recently, data-driven models have been used widely to estimate the scour depth around bridge piers and abutments. Najafzadeh, Barani and Hessami-Kermani (2015) used a set of data-driven models such as the gravitational search algorithm and the particle swarm optimization algorithm to estimate the scour around bridge abutments. Sediment particles size, geometric characteristics of bridge abutment and approach flow conditions were considered as effective parameters on scouring mechanism. The results of this research showed that data-driven models can estimate the scour depth with high accuracy. In the present study, M5 tree model has been used to predict maximum scour depth around bridge abutment. High accuracy, simplicity and understandability are considered the advantages of this algorithm. A huge amount of data from credible references have been used to build and validate the tree model.

کلیدواژه‌ها [English]

  • bridge abutment
  • M5 model
  • Pruning
  • Flow depth-abutment length
  • Sensitivity analysis

1-    Ballio, F., Teruzzi, A. and Radice, A., 2009. Constriction effects in clear-water scour at abutments. Journal of Hydraulic Engineering135(2), pp.140-145.

2-    Bayram, A. and Larson, M., 2000. Analysis of scour around a group of vertical piles in the field. Journal of waterway, port, coastal, and ocean engineering126(4), pp.215-220.

3-    Bhattacharya, B. and Solomatine, D.P., 2006. Machine learning in sedimentation modelling. Neural Networks19(2), pp.208-214.

4-    Coleman, S.E., Lauchlan, C.S. and Melville, B.W., 2003. Clear-water scour development at bridge abutments. Journal of Hydraulic Research41(5), pp.521-531.

5-    Dey, S., Chiew, Y.M. and Kadam, M.S., 2008. Local scour and riprap stability at an abutment in a degrading bed. Journal of Hydraulic Engineering134(10), pp.1496-1502.

6-    Dongol, D.M.S. and Melville, B.W., 1994. Local scour at bridge abutments. Department of Civil Engineering, University of Auckland.

7-    Etemad-Shahidi, A. and Ghaemi, N., 2011. Model tree approach for prediction of pile groups scour due to waves. Ocean Engineering38(13), pp.1522-1527.

8-    Etemad-Shahidi, A. and Mahjoobi, J., 2009. Comparison between M5′ model tree and neural networks for prediction of significant wave height in Lake Superior. Ocean Engineering36(15-16), pp.1175-1181.

9-    Etemad-Shahidi, A. and Taghipour, M., 2012. Predicting longitudinal dispersion coefficient in natural streams using M5′ model tree. Journal of hydraulic engineering138(6), pp.542-554.

10- Froehlich, D.C., 1989. Local scour at bridge abutments. In Proceedings of the 1989 National Conference on Hydraulic Engineering (pp. 13-18).

11- Gill, M.A., 1970. Bed erosion around obstructions in rivers (Doctoral dissertation, University of London).

12- Gill, M.A., 1972. Erosion of sand beds around spur dikes. Journal of the Hydraulics Division98(hy9).

13- Hager, W.H. and Oliveto, G., 2002. Shields’ entrainment criterion in bridge hydraulics. Journal of Hydraulic Engineering128(5), pp.538-542.

14- Kandasamy, J.K. and Melville, B.W., 1998. Maximum local scour depth at bridge piers and abutments. Journal of hydraulic research36(2), pp.183-198.

15- Kuhnle, R.A., Alonso, C.V. and Shields, F.D., 1999. Geometry of scour holes associated with 90 spur dikes. Journal of Hydraulic Engineering125(9), pp.972-978.

16- Kwan, T.F., Melville, B.W. and Raudkivi, A.J., 1984. Study of Abutment Scour: A Report Submitted to the Road Research Unit of the National Roads Board. Department of Civil Engineering, University of Auckland.

17- Laursen, E.M., 1962. Scour at bridge crossings. Transactions of the American Society of Civil Engineers127(1), pp.166-179.

18- Laursen, E.M., 1963. An analysis of relief bridge scour. Journal of the Hydraulics Division89(3), pp.93-118.

19- Melville, B.W. and Coleman, S.E., 2000. Bridge scour. Water Resources Publication.

20- Melville, B.W., 1992. Local scour at bridge abutments. Journal of Hydraulic Engineering118(4), pp.615-631.

21- Melville, B.W., 1997. Pier and abutment scour: integrated approach. Journal of hydraulic Engineering123(2), pp.125-136.

22- Najafzadeh, M., Barani, G.A. and Hessami-Kermani, M.R., 2015. Evaluation of GMDH networks for prediction of local scour depth at bridge abutments in coarse sediments with thinly armored beds. Ocean Engineering104, pp.387-396.

23- Quinlan, J.R., 1992, November. Learning with continuous classes. In 5th Australian joint conference on artificial intelligence (Vol. 92, pp. 343-348).

24- Richardson, E.V. and Davis, S.R., 1995. Evaluating scour at bridges: Federal Highway Administration Hydraulic Engineering Circular No. 18. Publication FHWA-IP-90-017.

25- Seo, I.W. and Cheong, T.S., 1998. Predicting longitudinal dispersion coefficient in natural streams. Journal of hydraulic engineering124(1), pp.25-32.

26- Sumer, B.M., Fredsøe, J. and Christiansen, N., 1992. Scour around vertical pile in waves. Journal of waterway, port, coastal, and ocean engineering118(1), pp.15-31.

27- Tey, C.B., 1984. Local scour at bridge abutments. Department of Civil Engineering, University of Auckland.

28- Wang, Y. and Witten, I.H., 1996. Induction of model trees for predicting continuous classes. Proceedings of the Poster Papers of the European Conference on Machine Learning, University of Economics, Faculty of Informatics and Statistics, Prague.

29- White, W.R., Milli, H. and Crabbe, A.D., 1973. Sediment transport: an appraisal methods, Vol. 2: Performance of theoretical methods when applied to flume and field data. Hydraulic Research Station Report No. IT119, Wallingford, UK.

30- Yorozuya, A. and Ettema, R., 2015. Three abutment scour conditions at bridge waterways. Journal of Hydraulic Engineering141(12), p.04015028.