نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی ارشد گروه علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه تبریز، تبریز، ایران

2 دانشیار، گروه علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه تبریز، تبریز، ایران

3 دانشجوی کارشناسی ارشد، گروه علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه تبریز، تبریز، ایران

4 دکترای مهندسی آب، کارشناس دفتر مطالعات منابع آب، شرکت آب منطقه‌ای استان قزوین، قزوین، ایران

چکیده

مدل‌های داده مبنا به‌عنوان یک جایگزین برای روش‌های هیدرولوژیکی در محاسبات مربوط به تخمین رسوب مطرح هستند. هدف پژوهش حاضر مقایسه عملکرد و دقت روش‌های هیدرولوژیکی و داده- مبنا در برآورد میزان رسوب معلق بود. بدین منظور داده‌های دبی و رسوب در بازه زمانی yr 20 (1399-1380) جمع‌آوری و سپس میزان رسوب معلق ایستگاه هیدرومتری باغ کلایه بر روی رودخانه الموت در استان قزوین برآورد شد. در این پژوهش از روش‌های هیدرولوژیکی شامل منحنی سنجه رسوب، فائو و روش اصلاح‌گر و روش‌های داده-مبنای برنامه­ریزی بیان ژن، یادگیری بر پایه نمونه K و رگرسیون خطی استفاده شد. عملکرد روش‌های مذکور با معیارهای R، RRMSE و NS مورد ارزیابی قرار گرفت. نتایج نشان داد که به ترتیب روش یادگیری بر پایه نمونه K با معیارهای ارزیابی 94/0 R=، 29/0= RRMSE و 24/0= NS و روش برنامه‌ریزی بیان ژن با 85/0 R=، 59/0= RRMSE و 65/0= NS رسوب معلق را با دقت بیشتری نسبت به سایر روش‌های موردمطالعه برآورد کرده است. بدین ترتیب برتری روش­های داده-مبنا در برآورد میزان رسوب معلق در منطقه موردمطالعه به اثبات رسید. ازاین‌روی استفاده از روش­های داده-مبنا به‌عنوان رقیب و جایگزین روش­های هیدرولوژیکی برای تخمین میزان رسوب معلق در مناطقی شبیه با منطقه موردمطالعه توصیه می‌شود.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Performance Analysis of Hydrological and Data Based models in Estimation of Suspended Sediment Rate

نویسندگان [English]

  • Sahar Javidan 1
  • Mohammad Taghi Sattari 2
  • Paria Karimzadeh 3
  • Ahmad Mehrabi 4

1 M.Sc. Student, Department of Water Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran

2 Assoc. Professor, Department of Water Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran

3 M.Sc. Student, Department of Water Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran

4 PhD in Water Engineering, Expert in the Office of Water Resources Studies, Qazvin Regional Water Company, Qazvin, Iran

چکیده [English]

Data driven models are proposed as an alternative to hydrological methods in sediment estimation calculations. The aim of this study was to compare the performance and accuracy of hydrological and data-based methods in estimating the amount of suspended sediment. For this purpose, discharge and sediment data were collected in the period of 20 yr (2001-2011) and then the amount of suspended sediment of Bagh Kalayeh hydrometric station on Alamut River in Qazvin province was estimated. In this study hydrological methods including Smearing, FAO and Sediment Rating Curves versus data driven methods including Gene Expression Programming, Instance-Based Learning with parameter K and Linear Regression methods were used. The model performances were compared using two statistical methods of RRMSE and NS. The results showed that two techniques such as IBK model with evaluation criteria of (R = 0.94, RRMSE = 0.29 and NS = 0.24) and the GEP model with (R = 0.85, RRMSE = 0.59 and NS = 0.65) estimated suspended sediment in more accurate way than other studies methods. Thus, the superiority of data-driven methods in estimating the amount of suspended sediment in the study area was proved. Therefore, the use of data-based techniques as a competitor and alternative to hydrological methods to estimate the amount of suspended sediment in areas similar to the study area is recommended.

کلیدواژه‌ها [English]

  • Gene Expression Programming
  • Linear Regression
  • Sediment Rating Curves
  • Smearing Method
  • Suspended Sediment
Aha, D. W., Kibler, D. and Albert, M. K. (1991). Instance-based learning algorithms. Machine Learn., 6, 37-66.
Anonymous (2021). Geographical area of Qazvin city. Qazvin Municipality. https://www.qazvin.ir/geographical-area. [In Persian]
Aytek, A. and Kisi, O. (2008). A genetic programming approach to suspended sediment modelling. J. Hydrol., 351, 288-298. doi:10.1016/j.jhydrol.2007.12.005.
Dadashzadeh, F., Mohseni, M., Ahmadi, H. and varvani, J. (2009). Evaluation and developing sediment rating curves models for estimating sediment yield of flood events (Case of study: Ghare Chay Basin). Watershed Manage. Res., 84, 28-35 [In Persian].
Dehghani, A. A., Malek Mohammadi, M. and Hezarjaribi, A. (2010). Estimation of Suspended Sediment Load in Behesht Abad River by Using Artificial Neural Network. J. Water Soil Conserv., 17(1), 159-168 [In Persian].
Dehghani, N. and Vafakhah, M. (2013). Comparison of daily suspended sediment load estimations by sediment rating curve and neural network models (Case Study: Ghazaghli Station in Golestan Province). J. Water Soil Conserv., 20(2), 221-1230 [In Persian]
Emamgholizadeh, S., Karimi, R. and Azhdari, K. (2016). Comparison of conventional methods for estimating suspended sediment load of Karkheh river with gene expression planning method. Quart. J. Geogra. Develop., 45, 121-140. [In Persian]
Fathzadeh, A., Asadi, M. and Taghizadeh, R. (2017). Optimization of suspended load estimation models using morphological geometry parameters and feature reduction technique. Iran. Soil Water Res., 3, 669-678. DOI:10.22059/ijswr.2017.210038.667483 [In Persian].
Ferreira, C. (2001). Gene expression program-ming a new adaptive algorithm for solving problems. Complex Sys., 13(2), 87-129.
Iadanza, C. and Napolitano, F. (2006). Sediment transport time series in the Tiber River. Phys. Chem. Earth, 31, 1212-1227.
Jalali, V. R. and Homaee, M. (2011). Introducing a nonparametric model using k-nearest neighbor technique for predicting soil bulk density. J. Water Soil Sci., 15(56), 181-191.
Jones, K. R., Berney, O., Carr, D. P. and Barret, E. C. (1981). Arid zone hydrology for agricultural development. FAO Irrig. Drain. Paper, 37, 271-284.
Kargar, K., Sadegh Safari, J., Mohammadi, A. and Samadianfard, S. (2019). Sediment transport modeling in open channels using neuro-fuzzy and gene expression programming techniques. Water Sci. Technol., 79(12), 2318–2327. doi:10.2166/wst.2019.229
Keihani, A. R., Mohammadi, A. and Fathian, H. (2021). Uncertainty analysis of SVM model parameters for estimating suspended and bed sediment load at Sierra station in Karaj by Monte-Carlo simulation method. Iran. Soil Water Res., 52(1), 195-212. Doi: 10.22059/IJSWR.2020.308225.668704. [In Persian]
Khosravi, K., Mao, L., Kisi, O., Yaseen, Z. and Shahid, Sh. (2018). Quantifying hourly suspended sediment load using data mining models: case study of a Glacierized Andean Catchment in Chile. J. Hydrol., 567, 165-179. Doi: 10.1016/j.jhydrol.2018.10.015.
Pavanelli, D. and Bigi, A. (2004). Suspended sediment concentration for three apennine monitored basins, particle size distribution and physical parameters. Agro Environ. Congress, Venice, Italy, 537 -544.
Ramezanipour, E., Mosaedi, A. and Mesdaghi, M. (2017). Determination of the best model for estimation of suspended sediment by using statistical error criteria (Case study: some sub-watersheds of Kashaf Roud). J. Watershed Res., 8(15), 112-124. Doi:10.29252/jwmr.8.15.112 [In Persian].
Rencher Alvin C. and Christensen William F. (2012). Multivariate regression. (3rd ed.), John Wiley & Sons, p. 19, ISBN 978-1-118-39167-9.
Rezazadeh, A. and Sattari., M. T. (2017). Comparison of the efficiency of support vector regression methods and k-nearest neighborhood in estimating the amount of suspended sediment load in the river (Case study: Ligvan Chai River). Iran. J. Nat. Resour., 2, 345-358 [In Persian].
Roushangar, K. and Shahnazi, S. (2019). Evaluating the performance of data-driven methods for prediction of total sediment load in gravel-bed rivers. Iran. Soil Water Res., 50(6), 1467-1477. Doi: 10.22059/ijswr.2019.253848.667867 [In Persian].
Saghebian, M. (2021). Estimation of sediment suspended load using integrated intelligent methods taking into account model uncertainty. J. Water Soil, Doi:10.22067/JSW.2021.68665.1021 [In Persian].
Shadkani, S., Abbaspour, A., Samadianfard, S., Hashemi, S., Mosavi, A. and Band, S.S. (2020). Comparative study of multilayer perceptron-stochastic gradient descent and gradient boosted trees for predicting daily suspended sediment load: the case study of the Mississippi River. Int. J. Sediment Res., 36(4), 512-523. Doi.org/10.1016/j.ijsrc.2020.10.001.
Shirzad, A., Soltani, F., and Zare Abyaneh, H. (2008). Simulation of scouring in accordance with energy-depleting structures using k-nearest closest neighbor (KNN) algorithm and non-fuzzy adaptive inference system (ANFIS). First International Water Crisis Summit, University of Zabol [In Persion]
Warrick, A. W. and Nielsen, D. R. (1980). Spatial variability of soil physical properties in the field. In: Encycl. Agrophys., 319-344. Doi: 10.1007/978-90-481-3585-1_163
Wu, X., Kumar, V., Quinlan, J. R., Ghosh, J., Yang, Q., Motoda, H., McLachlan, G. J., Ng A., Liu, B. and Philip, S. Y. (2008). Top 10 algorithms in data mining. Know. Inform. Syst., 14, 1–37. Doi.org/10.1007/s10115-007-0114-2.
Youssefvand, F. (2004). Suggestion of a method for estimation of suspended load in rivers (case study: Ghresoo river). M. Sc. Dissertation, Sari Agricultural Sciences and Natural Resources University. 138 pp. [In Persian]
Zounemat-Kermani, M., Mahdavi-Meymand, A., Alizamir, M., Adarsh, S. and Yaseen, Z. (2020). On the complexities of sediment load modeling using integrative machine learning: Application of the great river of Loíza in Puerto Rico. J. Hydrol., 585, 124759. Doi.org/10.1016/j.jhydrol.2020.124759.