نشریه علمی - پژوهشی مرتع و آبخیزداری

نوع مقاله : مقاله پژوهشی

نویسندگان

گروه مهندسی مرتع و آبخیزداری، دانشکده منابع طبیعی، دانشگاه لرستان، لرستان، ایران

10.22059/jrwm.2023.351372.1684

چکیده

این تحقیق با استفاده از الگوریتم‌های یادگیری ماشین به بررسی کارآیی مدل‌های RF, RepTree, GP-PUK, GP-RBF, M5P برای مدل‌سازی بارانحلالی در زیرحوضه‌های خرم‌آباد، بیرانشهر و الشتر در استان لرستان پرداخته شد. داده‌های ورودی شامل بارش، دبی، دبی یک روز قبل، میانگین دبی (دبی همان روز و یک روز قبل) همچنین داده خروجی رسوب انحلالی رودخانه‌ها می‌باشد. در این تحقیق برای مدل‌سازی در مرحله آموزش 70 درصد داده‌ها و در مرحله آزمایش 30 درصد باقی‌مانده مورد استفاده قرار گرفتند. در نهایت برای مقایسه نتایج مدل‌های مختلف و انتخاب بهترین مدل، از معیارهای سنجش خطای ریشه میانگین مربعات خطا (RMSE)، ضریب همبستگی (C.C) و میانگین مربعات خطا (MAE) استفاده شد. نتایج نشان داد باتوجه به معیارهای ارزیابی مدل GP با دو تابع کرنل PUK و RBF در دوره‌‌ پرآبی و کم‌آبی عملکرد بهتری را نسبت به سایر مدل‌ها داشته است. نتایج به‌دست آمده در دوره پرآبی نشان داد که در ایستگاه‌های چم‌انجیر، سراب صیدعلی و کاکارضا مدل GP-RBF و در ایستگاه هیدرومتری بهرام‌جو مدل GP-PUK با بیشترین ضریب همبستگی و کمترین خطا در مرحله آزمایش به‌عنوان مدل‌های بهینه برای تخمین بار انحلالی انتخاب شدند. همچنین در ایستگاه‌های هیدرومتری بهرام‌جو، چم‌انجیر و سراب صیدعلی مدل GP-RBF و در ایستگاه هیدرومتری کاکارضا مدل GP-PUK به‌عنوان مدل بهینه برای تخمین بار انحلالی در دوره کم‌آبی انتخاب شدند. بنابراین، با توجه به نتایج به دست آمده، می‌توان برای مدیریت کیفیت و کمیت منابع آب سطحی از مدل‌های بهینه GP-PUK و GP-RBF برای تخمین بار انحلالی رودخانه‌های فاقد ایستگاه هیدرومتری در حوضه‌های کارستی استفاده کرد.

کلیدواژه‌ها

عنوان مقاله [English]

Total Dissolved Solids modeling using machine learning algorithms in periods of low and high water (Case study: Khorammabad, Biranshahr and Alashtar watersheds, Lorestan province)

نویسندگان [English]

  • Nasrin Beiranvand
  • Alireza Sepahvand
  • Ali Haghizadeh

Department of Range and Watershed Management, Faculty of Natural Resources, Lorestan University, Lorestan, Iran.

چکیده [English]

In this study, five soft computing techniques, GP-PUK, GP-RBF, M5P, REEP Tree and RF were used to predict the SL in Cham Anjir, Bahram Joo, Kaka Reza and Sarab Syed Ali hydrometry stations in Khorramabad, Biranshahr and Alashtar sub-watersheds, Lorestan province. Total data set consists of rain, discharge and solute load (SL) of three sub-watersheds out of which 70% data used to training and 30% data were used to testing phase. Finally, the models’ accuracy was assessed using three performance evaluation parameters, which were Correlation Coefficient (C.C.), Root Mean Square Error (RMSE) and Maximum Absolute Error (MAE). Results suggest that GP-PUK and GP-RBF models works well than other modeling approaches in estimating the SL in low and high water-periods. The result showed that, In the high-water period, in Cham Anjir, Sarab Said Ali and Kaka Reza stations the GP-RBF model and in the Bahram Joo station the GP-PUK model with the highest C.C and the lowest error were selected the optimal models in estimating the SL. Also, in the low water period, result shown that in Cham Anjir, Sarab Said Ali and Bahram Joo stations the GP-RBF model and in the Kaka Reza station the GP-PUK model were the best models in estimating the SL. Therefore, these models can be used to estimate the solute load of nearby rivers by/without hydrometry station for the management of the quantity and quality of surface water.

کلیدواژه‌ها [English]

  • Lorestan province
  • Kashkan Watershed
  • Total Dissolved Solids (TDS)
  • Flow Duration Curve (FDC)
  • Gaussian process
  • Random Forest
Ansari, H., and Davari, K. (2008). Classification of dry period using standard precipitation index in GIS environment. Geographical Research, 39(60), 98-108 (In Persian).
Asadi, M., Fathzadeh, A., Kerry, R., Ebrahimi Khusfi, R. and Taghizadeh Mehrjardi, R. (2021). Prediction of river suspended sediment load using machine learning models and geo-morphometric parameters. Arabian Journal of Geoscience, 14(1926), https://doi.org/10.1007/s12517-021-07922-6. (In Persian).
Baghdar dokht, Z., Hidari, M., Ghobadi, M.H. and Shafei, A. (2006). Study of karst formations solubility in Kangir Dam site, 4th Iranian Conference on Geology Engineering and Environment, 621-637. (In Persian).
Bazuhair, S.A., Gohani, A. and Sen, Z. (1997). Determination of monthly wet and dry periods in Saudi Arabia. International Journal of Climatology, (17), 303-311.
Bonakdar, L. and Etemad Shahidi, A. (2011). Predicting wave run-up on rubble-mound structures using M5 model tree. Ocean Engineering, (38), 111-118. (In Persian).
Breiman, L. (1996). Bagging predictors. Machin Learning, 24(2), 123–140.
Chahouki, Z., Salajegheh, A., Mahdavi, M., Khalighi, Sh. And Asadi, S. (2013). Regional duration curve in Arid regions for ungauged basins. Journal of Range and Watershed management, 66(2), 251.265(In Persian).
Cobaner M. 2011. Evapotranspiration estimation by two different neuro-fuzzy inference systems. Journal of Hydrology, 398: 292–302.
Eshghi, P., Farzad Mehr, J., Dasturani, M.T. and Arabasadi, Z. (2015). Investigating the efficiency of intelligent models in estimating river suspended sediments. Watershed Management Research Journal, 7(14), 88-96. (In Persian).
Ewusi, A., Ahenkorah, I. and Aikins, D. 2021. Modelling of total dissolved solids in water supply systems using regression and supervised machine learning approaches. Appl Water Science, 11(13), https://doi.org/10.1007/s13201-020-01352-7
Ford, D.C. and Williams, P. (2007) Karst Hydrogeology and Geomorphology. John Wiley, Chichester, 562. https://doi.org/10.1002/9781118684986.
Güldal V. and Tongal, H. 2010. Comparison of recurrent neural network, adaptive neuro-fuzzy inference system and stochastic models in Egirdir Lake level forecasting. Water Resources Management, 24(1): 105–128.
Khosravi, K., Pham, B.T., Chapi, K., Shirzadi, A., Shahabi, H., Revhaug, I., Prakash, I. and Bui, D.T. (2018). A comparative assessment of decision trees algorithms for flash flood susceptibility modeling at Haraz watershed, northern Iran. Science of the Total Environment, 627, 744–755.
Kohestani, V.R., Hasanlorad, M. and Bazargan Lari, M. (2016). Prediction of the ultimate bearing capacity of surface foundations located on granular soils using the M5P tree model. Ferdowsi Civil Engineering, 2(27), 99-110 (In Persian).
Mahdavi, M. (2007). Applied hydrology. Tehran University Press, 357 pp. (In Persian).
Melesse, A.M., Khosravi, K., Tiefenbacher, J.P., Heddam, S., Kim, S., Mosavi, A. and Pham, B.T. (2020). River Water Salinity Prediction Using Hybrid Machine Learning Models. Water12(10), 2951. https://doi.org/10.3390/w12102951
Mirhashemi, SH., Panahi, M. and Zareei, L. (2020). Evaluation of M5P Algorithm for Estimation of Potential Evapotranspiration, Minimum and Maximum Temperature. Journal of Meteorology and Atmospheric Sciences, 2(4), 287-295(In Persian).
Montaseri, M. and Zaman Zad Ghavidel, S. (2017). Comparing the Performance of Artificial Intelligence Models in Estimating Water Quality Parameters in Periods of Low and High Water Flow, Journal of Water and Soil, 30(6), 1733-1747(In Persian).
Mostafazadeh, R. and Zabihi, M. (2016). Comparison of SPI, SPEI indices in meteorological drought assessment using R programming (study area: Kurdistan province). Journal of Earth and Space Physics, 42, 633-643 (In Persian).
Pal, M. (2005). Random forest classifier for remote sensing classification. International Journal of Remote Sensing, 26(1), 217-222.
Pal, M. and Deswal, S. (2010). Modelling pile capacity using Gaussian process regression. Computer. Geotechnical, 37, 942-947.
Prasad, A.M., Iverson, L.R. and Liaw, A. (2006). Newer classification and regression tree techniques: Bagging and random forests for ecological prediction. Ecosystems, 9(2), 181-199.
Quinlan, J.R. (1992). Learning with continuous classes. in Proceedings of the 5th Australian joint Conference on Artificial Intelligence. Hobart 16-18 November, 343-348.
Samadianfard, S., Salarifar, M., Javidan, S. and Mikaeili, F. (2020). Estimation of Daily Reference Evapotranspiration in Humid Climates Using Data-Driven Methods of Gaussian Process Regression, Support Vector Regression and Random Forest. Environment and Water Engineering, 6(4), 360-373(In Persian).
Sengorur B., Dogan E., Koklu R. and Samandar A. 2006. Dissolved oxygen estimation using artificial neural network for water quality control. Fresenius Environmental Bulletin, 15: 1064–1067.
Sepahvand A., Nazari Samani, A.A., Mohammadian, H., Ahmadi, H. and Feiz Nia, S. (2020). Seasonal variation of the solute and determine the solubility of limestone formations. Iranian Journal of Watershed Management Science and Engineering, 14(48), 21-32. (In Persian).
Sepahvand, A., Prelovsek, M., Nazari Samani, A. and Wasson, R.J. (2021). Solute transport and solutional denudation rate of carbonate karst in the semi-arid Zagros region (southwestern Iran). Journal of Cave and Karst Studies, 83(3), 93-108.
Sepahvand, A., Singh, B., Sihag, P., Nazari, A., Hasan Ahmadi, S. and Fiz Nia, S. (2019). Assessment of the various soft computing techniques to predict sodium absorption ratio (SAR). ISH Journal of Hydraulic Engineering, 27, 124-135.
Sepehvand, A. and Azizi Najafkali, Z. (2019). Suspended sediment modeling using Gaussian process and multi-layer perceptron models 15th National conference on Watershed Management Sciences and Engineering of Iran, 15, 1-6. (In Persian).
Sighn K.P., Basant A., Malik A. and Jain G. 2009. Artificial neural network modeling of the river water quality-A case study. Ecological Modelling, 220: 888–895
Sihag P., Singh V.P., Angelaki A., Kumar V., Sepahvand A. and Golia E. (2019). Modelling of infiltration using artificial intelligence techniques in semi-arid Iran, 64(13): 1647-1658.
Smakhtin, V.U. (2001). Low-flow hydrology: a review. Journal of Hydrology. 240, 147-186.
Soleimani, L., Derikund, B. and Sepehvand, A. (2021). Permeability modeling in different classes of soil texture using learning algorithms. Watershed Researches, 35(4), 1-150 (In Persian).
Sun, Z., Guo, H., Li, X., Lu, L. and Du, X. (2011). Estimating urban impervious surfaces from Landsat-5 TM imagery using multilayer perceptron neural network and support vector machine. Journal of Applied Remote Sensing, 5(1), 053501.
Wang, Y. and Witten, I.H. (1997). Inducing model trees for continuous classes. Proceedings of the Ninth European Conference on Machine Learning. Prague, Czech Republic: Springer. 13 pp.
Yang, D., Zhang, X., Pan R., Wang, Y. and Chen, Z. (2018). A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve. Journal of Power Sources, 384, 387-39.