پیش‌بینی بار معلق رودخانه‌ای با استفاده از شبکه عصبی مصنوعی و الگوریتم‌ ژنتیک مرتب‌سازی غیرغالب

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی دکتری آبیاری و زهکشی، گروه مهندسی آب، دانشگاه ارومیه

2 دانشیار گروه مهندسی آب، دانشگاه ارومیه، ارومیه، ایران

3 دانشیار، گروه مهندسی آبیاری و آبادانی، دانشکده مهندسی و فناوری کشاورزی، دانشگاه تهران، کرج، ایران.

چکیده

اطلاع از میزان فرسایش خاک و تولید رسوب، ویژگی‌های هواشناسی، ویژگی‌های هیدرولوژیکی رودخانه همانند دبی و همچنین عوامل انسانی، غالباً بسیار پیچیده، غیر‌قطعی و غیرخطی می‌باشند. لذا بکارگیری الگوریتم‌های هوش ماشینی (نظیر الگوریتم‌های یادگیری ماشین) گزینه مناسبی در شبیه‌سازی و پیش‌بینی متغیرهای کیفی آب رودخانه نظیر بار معلق تلقی می‌شود. هدف پژوهش حاضر، ارائه یک روش پیشنهادی برمبنای شبکه عصبی مصنوعی پرسپترون چندلایه و الگوریتم ژنتیک مرتب‌سازی غیرغالب برای پیش‌بینی بار معلق رودخانه‌ای می‌باشد. در روش پیشنهادی به­منظور آموزش شبکه عصبی مصنوعی پرسپترون چندلایه از روش پس‌انتشار خطا و تعیین وزن بهینه برای نرون‌ها از  الگوریتم ژنتیک مرتب‌سازی غیرغالب استفاده شد. در این مطالعه از بار معلق ایستگاه تیل‌آباد واقع در رودخانه گرگان‌رود طی سال‌های 94-1361 به‌عنوان مطالعه موردی استفاده شد. نتایج نشان داد که روش پیشنهادی در مقایسه با شبکه عصبی مصنوعی چندلایه دارای ضریب همبستگی بالاتری است و مقدار R2 به‌ترتیب برابر با 6728/0 و 4372/0 بدست آمد.  مقدار RMSE در روش پیشنهادی و شبکه عصبی مصنوعی چندلایه برمبنای الگوریتم پس‌انتشار به­ترتیب برابر با 7225/4 و 0548/8 بدست آمده است. مقدار NSE نیز در روش پیشنهادی و شبکه عصبی مصنوعی چندلایه برمبنای الگوریتم پس‌انتشار به­ترتیب برابر با 4321/0 و 2941/0 بدست آمده است. لذا در روش پیشنهادی، الگوریتم‌ ژنتیک مرتب‌سازی غیرغالب باعث شده که شبکه عصبی مصنوعی چندلایه بهبود خوبی داشته باشد. نتایج حاصله نشان داد که روش پیشنهادی دارای دقت خوبی در پیش‌بینی بار معلق بوده است. روش پیشنهادی با الگوریتم آموزشی پس‌انتشار دارای عملکرد بهتری در مقایسه با الگوریتم آموزشی گرادیان نزولی و بیزین بوده است.

کلیدواژه‌ها

عنوان مقاله [English]

Predicting River Suspended Load Using Artificial Neural Network and Non-Dominant Genetic Sorting Algorithm

نویسندگان [English]

  • Mikael Hosseini 1
  • Mohammad Hemmati 2
  • Mehdi Yasi 3
1 Ph.D. Student of Department of Water Engineering, Urmia University
2 Associate professor, water engineering, Urmia University, Urmia, Iran
3 Associate Professor, Department of Irrigation and Reclamation Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, Karaj, Iran.
چکیده [English]

Information on soil erosion and sediment production, meteorological features, hydrological features of rivers such as discharge, as well as human factors, are often very complex, indefinite, and nonlinear. Therefore, the use of machine intelligence algorithms (such as machine learning algorithms) is a good option in simulating and predicting river water quality variables such as suspended load. The aim of the present study is to present a proposed method based on Multilayer Perceptron Artificial Neural Network (MLP-ANN) and Non-Dominant Sorting Genetic Algorithm (NSGA) for predicting suspended river load. In the proposed method, the NSGA was used to train the MLP using the error propagation method and determining the optimal weight for the neurons. In this study, the suspended load of Tilabad station located in Gorganrood river during the 1982-2015 years was used as a case study. The results showed that the proposed method has a higher correlation coefficient compared to MLP and the value of R2 was 0.6728 and 0.4372, respectively. The value of Root-Mean-Square Error (RMSE) in the proposed method and MLP based on back-propagation (BP) algorithm is 4.7225 and 8.548, respectively. Therefore, in the proposed method, the NSGA has caused a good improvement of the MLP. The NSE value in the proposed method and MLP based on BP algorithm is 0.4321 and 0.2941, respectively. The results showed that the proposed method had good accuracy in predicting the suspended load. The proposed method with BP training algorithm has a better performance compared to the descending and Bayesian gradient training algorithm.

کلیدواژه‌ها [English]

  • Sediment Rating Curve
  • Non-dominant Sorting Genetic Algorithm
  • Gorganrood River
  • Suspended Load

مراجع

Abdi Dehkordi M., Dehghani A.A., Meftah Halghi M., Kahe M., Hesam M., and Dehghani, N. 2014. Use of fuzzy clustering algorithm for estimating the daily suspended load (Case Study: Kasilain Basin). Journal of Water and Soil Science, 18(68): 121-131. (In Persian)
Ak R., Li Y., Vitelli V., Zio E., Droguett E.L., Jacinto C.M.C. 2013. NSGA-II-trained neural network approach to the estimation of prediction intervals of scale deposition rate in oil & gas equipment. Expert Systems with Applications, 40(4): 1205-1212.
Alijanpour Shalmani A., Vaezi A., and Tabatabaei M. 2020. Optimal combinations of hydrological variables for modeling of daily suspended sediment load in Karaj Watershed. Watershed Engineering and Management, 12(1): 228-243. (In Persian)
Asadi M., Fathzadeh A., and Taghizadeh Mehrjerdi R. 2017. The effects of the daily, monthly, and annual time scales on the suspended sediment load prediction. Hydrogeomorphology, 4(10): 121-143. (In Persian)
Banadkooki F.B., Ehteram M., Ahmed A.N., Teo F.Y., Ebrahimi M., Fai C.M., and El-Shafie, A. 2020. Suspended sediment load prediction using artificial neural network and ant lion optimization algorithm. Environmental Science and Pollution Research, 27(30): 38094-38116.
Barzegari F., and Dastorani M.T. 2016. Suspended sediment prediction using time series and artificial neural networks models (Case Study: Ghazaghly Station in Gorganroud River). Journal of Watershed Management Research, 6(12), 216-225. (In Persian)
Dastorani M.T., Azimi Fashi K., Talebi A., and Ekhtesasi M.R. 2012. Estimation of suspended sediment using artificial neural network (Case Study: Jamishan Watershed in Kermanshah). Journal of Watershed Management Research, 3(6): 61-74. (In Persian)
Emami S., Choopan Y., and Parsa J. 2021. Comparising performance of meta-heuristic algorithms with the sediment rate curve (Case Study: Zarrineh Rood River). Watershed Engineering and Management, 13(1): 43-54. (In Persian)
Gharehchopogh F.S., and Gholizadeh H. 2019. A comprehensive survey: Whale Optimization Algorithm and its applications. Swarm and Evolutionary Computation, 48: 1-24.
Gupta D., Hazarika B.B., Berlin M., Sharma U.M., and Mishra K. 2021. Artificial intelligence for suspended sediment load prediction: a review. Environmental Earth Sciences, 80(9): 346.
Halabian A., Javari M., Akbari Z., and Akbari G. 2017. Evaluating the performance of decision tree model in estimating the suspended sediments of river (A case study on the Basin of Meimeh River). Geography And Development Iranian Journal, 15(49): 81-96. (In Persian)
Hayatzadeh M., Chezgi J., and Taghi Dastorani M. 2015. Evaluation of sediments using rating curve and artificial neural network methods by combining morphological parameters of basin (case study: bagh abbas basin). Journal of Water and Soil Science, 19(72): 217-228. (In Persian)
Imamgholizadeh S., Karimi Demneh R., and Azhdari K. 2016. Comparison of conventional methods for estimation of suspended sediment load of karkheh river by gene expression programming approach. Geography And Development Iranian Journal, 14(45): 121-140. (In Persian)
Keihani A., Akhondali A.M., 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. Iranian Journal of Soil and Water Research, 52(1): 195-212. (In Persian)
Khazaei Moughani S., Najafinejad A., Azimmohseni M., and Berdi Sheikh V. 2013. Spatial and seasonal variation of suspended sediment in different stations of gorganroud river. golestan province. Journal of Watershed Management Research, 4(7): 1-15. (In Persian)
Khazaie Poul A., and Talebi A. 2013. Investigation of possibility of suspended sediment prediction using a combination of sediment rating curve and artificial neural network (case study: ghatorchai river, yazdakan bridge). Quarterly Journal of Environmental Erosion Researches, 3(1): 73-82. (In Persian)
Kheirfam H., and Mokarram-Kashtiban S. 2018. A regional suspended load yield estimation model for ungauged watersheds. Water Science and Engineering, 11(4): 328-337
Khorshiddoost A., Feyzolahpour M., and Sadrafshary S. 2015. Assessing the capability of adaptive neuro fuzzy interference system (anfis) in estimating the amount of suspended sediment load and its comparison with two models of artificial neutral fuzzy inference system case study: zarine rood, south east basin of urmia lake. Geography And Development Iranian Journal, 13(41): 185-200. (In Persian)
Kia E., Emadi A., and Fazlola R. 2015. Investigation for application of adaptive neuro-fuzzy inference system (anfis) in babolroud suspended sediment load estimation. Journal of Watershed Management Research, 6(11): 15-23. (In Persian)
Najafinejad A., Mardian M., Varvani J., and Sheikh V.B. 2011. Performance evaluation of correction factors in optimization of sediment rating curve (case study: kamal saleh dam watershed, markazi province, iran). Journal of Water and Soil Conservation,  18(2): 105-122.
Nikkami D., and Shadfar S. 2021. Soil erosion mapping in sediment gauged watersheds of Iran. Watershed Engineering and Management, 13(2): 479-496.
Panahi F., Ehteram M., and Emami M. 2021. Suspended sediment load prediction based on soft computing models and Black Widow Optimization Algorithm using an enhanced gamma test. Environmental Science and Pollution Research, 28(35): 1-21.
Rahgoshay M., Feiznia S., Arian M., and Hashemi S.A.A. 2019. Simulation of daily suspended sediment load using an improved model of support vector machine and genetic algorithms and particle swarm. Arabian Journal of Geosciences, 12(9): 277.
Rahmani V., Kastens J. H., DeNoyelles F., Jakubauskas M.E., Martinko E.A., Huggins D.H., Blackwood A.J. 2018. Examining storage capacity loss and sedimentation rate of large reservoirs in the central u.s. great plains. Water, 10(2): 1-17
Rajaee T., Mirbagheri S. A., Nourani V., & Alikhani A. 2010. Prediction of daily suspended sediment load using wavelet and neurofuzzy combined model. International Journal of Environmental Science & Technology, 7(1): 93-110.
Razavizadeh S., Kavian A., and Vafakhah M. 2014. Estimation of suspended sediment discharge by optimal structure of artificial neural network in taleghan watershed. Journal of Water and Soil Science, 18(68): 79-88. (In Persian)
Shafaghat R., Hosseinalipour S., Mohammad Nouri N., and Lashgari I. 2008. Shape optimization of two-dimensional cavitators in supercavitating flows, using NSGA II algorithm. Applied Ocean Research, 30(4): 305-310.
Tabatabaei M., Salehpour Jam A., and Mosaffaie J. 2020. Improvement of the efficiency of artificial neural network model in suspended sediment simulation using particle swarm optimization algorithm. Watershed Engineering and Management, 12(3): 756-770. (In Persian)
Tabatabaei M.R, Solaimani K., Habibnejad Roshan M., and Kavian A. 2014. Estimation of daily suspended sediment concentration using artificial neural networks and data clustering by self-organizing map (case study: sierra hydrometry station- karaj dam watershed). Journal of Watershed Management Research, 5(10): 98-116. (In Persian)
Varvani J., Gholi Ebrahimi N., Yosefi M., Shadmani A.R., and Nikche Farahani S. 2016. A performance evaluation of neuro-fuzzy and regression methods in estmation of sediment load of river. Watershed Management Research (Pajouhesh & Sazandegi), 28(4): 104-113. (In Persian)
Yilmaz B., Aras, E., Kankal M., and Nacar S. 2019. Prediction of suspended sediment loading by means of hybrid artificial intelligence approaches. Acta Geophysica, 67(6): 1693-1705.
Yosefi M., and Poorshariaty R. 2015. Suspended sediment estimation using neural network and algorithms assessment (case study: lorestan province). Journal of Watershed Management Research, 5(10): 85-97. (In Persian)
Yousefi M., and Barzegari F. 2014. Estimation of suspeded sediment using decision tree method and copmarision the results with rating curve method (Case study: Rivers in Lorestan Province). Watershed Management Research (Pajouhesh & Sazandegi), 27(3): 187-195. (In Persian)
Zare M., Pourghasemi H.R., Vafakhah M., and Pradhan B. 2013. Landslide susceptibility mapping at Vaz Watershed (Iran) using an artificial neural network model: a comparison between multilayer perceptron (MLP) and radial basic function (RBF) algorithms. Arabian Journal of Geosciences, 6(1): 2873-2888.
Zhao G., Pang B., Xu Z., and Xu L. 2020. A hybrid machine learning framework for real-time water level prediction in high sediment load reaches. Journal of Hydrology, 581(1): 1-10.

فایل ها

سابقه مقاله

  • تاریخ دریافت: 04 مرداد 1400
  • تاریخ بازنگری: 10 مهر 1400
  • تاریخ پذیرش: 15 مهر 1400

هم رسانی

ارجاع به این مقاله

آمار