Application of M5 Model for Rivers Suspended Sediment Evaluation

Document Type : Original Article

Authors

1 M. Sc. Water structures, Khuzestan Ramin Agriculture and Natural Resources University

2 Assistant Professor, Department of Water Engineering, Khouzestan Ramin Agriculture and Natural Resources University

Abstract

Estimation of the suspended sediment load plays a very important role in water projects such as dam and surface water storage design, pollution and sediment transport in rivers and lakes, as well as design and maintenance the channels. The accuracy of conventional methods of sediment estimation, usually low and there are a big difference between their results. Therefore, in this study after preparing sediment and flow discharge data of hydrometric stations; Jow Kanak, Behbahan, Shadegan, Moshrageh and Cham Nezam, the M5 algorithm, has been used to determine the suspended sediment load. M5 tree algorithm is one of the newest data-driven models which divides the problem space into multiple branches and proposes equations for each branches. Flow discharge of each day and the day before have been used as an input variables and sediment discharge has been considered as an output variable to build and validate the tree model. Statistical analysis on tree model showed better compliance of tree model results with observed data in compare with sediment rating curve method. RMSE for training and validation phases at studied stations were 0.59 and 0.74 for M5 and rating curve methods, respectively. The results of this study showed that suspended sediment load could be accurately predicted using M5 tree algorithm. The results of this model, in addition to have an understandable and simple structure, could be used in practical issues as well.

Keywords

References

 
اکبرزاده، م. ر.، ع. قزل سوفلو و م. حاجی بیگلو. 1393. ارزیابی و برآورد رسوب رودخانه اترک در محل ایستگاه‌های هیدرومتری دارای آمار با استفاده از روش‌های هیدورلوژیکی. مجله علوم و مهندسی آبیاری، دوره 37، شماره 4، ص 132-121.
جباری، ا و م. صمدی. 1392. کاربرد الگوریتم M5 در پیش بینی عمق آبشستگی در پایین‌دست سرریزها. طرح تحقیقات کاربردی، شرکت سهامی مدیریت منابع آب ایران.
دهقانی، ا.، م. ملک‏ محمدی و ا. هزارجریبی. 1389. تخمین بار معلق رودخانه بهشت آباد با استفاده از شبکه عصبی مصنوعی. مجله پژوهش‏های حفاظت آب و خاک، شماره 17، ص 168-159.
ابراهیمی ‏محمدی، ش.، س. ح. ر. صادقی و ک. چپی. 1391. تحلیل آورد رواناب، رسوب معلق مغذی ورودی‏های مختلف به دریاچه زریوار در پایه زمانی رگبار و آب پایه. نشریه حفاظت منابع آب و خاک، شماره 2، ص 74-61.
ظهیری، ج. 1394. کاربرد مدل‏های ناپارامتریک CART و M5 در محاسبه عمق آبشستگی اطراف پایه‏های پل. فصلنامه علمی پژوهشی مهندسی آبیاری و آب ، شماره 20، ص 50-35.
معتمدی‏وزیری، ب.، ح. احمدی، م. مهدوی و ف. شریفی. 1388. مقایسه روش‏های رگرسیون آماری و فازی جهت برآورد رسوب معلق رودخانه تلوار. فصلنامه جغرافیایی سرزمین علمی پژوهشی، شماره 21، ص 42-31.
Achite, M. and S. Ouillon. 2007. Suspended sediment transport in a semiarid watershed, Wadi Abd, Algeria (1973–1995). Journal of Hydrology, 343: 187-202.
Bhattacharya, B. and D. P. Solomatine. 2006. Machine Learning in Sedimentation Modeling. Neural Networks Journal, 19: 208–214.
Etemad-Shahidi, A. and J. Mahjoobi .2009. Comparison between M5΄ Model Tree and Neural Networks for prediction of significant wave height. Journal of Ocean Engineering, 36(15-16): 1175-1181.
Etemad-Shahidi, A. and M. Taghipour. 2012. Predicting Longitudinal Dispersion Coefficient in Natural Streams Using M5′ Model Tree. Journal of Hydraulic Engineering, 138(6): 542-554.
Etemad-Shahidi, A. and N. Ghaemi. 2011. Model Tree Approach for Prediction of Pile Groups Scour due to Waves. Ocean Engineering, 38: 1522–1527.
Kumar, D., A. Pandey, N. Sharma and W. Flugel .2016. Daily suspended sediment simulation using machine learning approach. Catena, 138: 77-90.
Mahjoobi, J., A. Sabzianpoor and E. Jabbari. 2010. Application of Meta-Heuristic Models for Local Scour Evaluation. Ninth International Conference on Computing Anticipatory Systems, AIP Conference Proceedings, 1303:389-397.
Quinlan, J. R.1992. Learning with Continuous Classes. Proceedings of AI’92, World Scientific, 343–348.
Wang, Y. and I. H. Witten. 1997. 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.
Yang, C. T., R. Marsooli and M. Taghi Aalami. 2009. Evaluation of total load sediment transport formulas using ann. International Journal of Sediment Research, 24: 274-286.
Zounemat-Kermani, M., O. Kişi, J. Adamowski and A. Ramezani-Charmahineh. 2016. Evaluation of data driven models for river suspended sediment concentration modeling, Journal of Hydrology, 535: 457-472. 
Irrigation and Water Engineering
Volume 6, Issue 4 - Serial Number 24
September 2016
Pages 16-28

Files

Share

How to cite

Statistics