Compare Learning Function in Neural Networks for River Runoff Modeling

Document Type : Research Article


1 Ph.D. Student. Department of Science and Water Engineering University of Birjand

2 Department of Science and Water Engineering University of Birjand


Accurate prediction of river flow is one of the most important factors in surface water recourses management especially during floods and drought periods. In fact deriving a proper method for flow forecasting is an important challenge in water resources management and engineering. Although, during recent decades, some black box models based on artificial neural networks (ANN), have been developed to overcome this problem and the accuracy privilege to common statistical methods (such as auto regression and moving average time series method) have been shown. In these research only attended change number of hidden layer and number of neurons for to approach to the best structure in neural network, and complex in proper network designand can’t be simply used by other investigators. In this study examined 15 the neural network learning function and the results indicated in the structure of the network with one hidden layer (ANN1),learnlv1, learnh and learnis by MSE=0.000158, 0.000185 and 0.000188, have been better performance than the other learning functions. And in the structure of the network with two hidden layer (ANN2),learnh, learnsomb and learncon learning function by MSE=0.000154, 0.000173 and 0.000176 have been better performance than the other learning functions.But on the other hand by ten times run this two models, learnsom and learngdm learning functions in ANN1 model and learnh and learnos in ANN2 model had most frequency among the best learning functions and thus it is better that the number of hidden layer not more than one, when we use back propagation network (that its learning function is learngdm). Because in this way we have more chance to achieve ideal response. But if we are going to increase network performance byincreasing the number of hidden layer, it is better that use the default of network and learngdm carefully.


Main Subjects

[1].Tokar AS, Markus M. Precipitation – runoff modeling using artificialneural network and conceptual models. Journal of Hydrologic Engineering. 2000;4:150-161.
[2].Razavi SS, Karamuoz M. in Prediction monthly river flows by using artificial aeural network. 10th studentsConferenceonCivil Engineering. Amirkabir University of Technology. 22 Oct 2003. [Persian]
[3].Fathi P, Mohammadi Y, Homayi M. Intelligent modeling of monthly flow time series into vahdat dam in sanandaj city. Journal of Water and Soil. 2009; 23(1):209-220. [Persian]
[4].Dorum A, Yarar A, FaikSevimli M, and Onucyildiz M. Modelling the rainfall-runoff data of susurluk basin. Expert Systems with Applications. 2010. 37: 6587-6593.
[5].Chua HC, and Wong SW. Rainfall-runoff modeling using artificial neural network coupled with singular spectrum analysis. JournalHydrology. 2011. 399: 394-409.
[6].Patil S,Valunjkar, S. Study of different rainfall-runoff forecasting algorithms for better water consumption. International Conference on Computational Techniques and Artificial Intelligence. 2012. 327-330.
[7].Zeynali MJ, Nikbakht S, Mohammadezapour O. Prediction Input Flows to Mollasadra Reservoir by Useing Artificial Neural Network. 5th Iranian water resources management conference. ShahidbeheshtiUniversity.29 jul 2013. [Persian]
[8].Braddock RD,Kremmer ML, Sanzogni L. Feedforward artificial neural network model forforecasting rainfall-runoff. Journal of Environmental Sciences. 1998. 9:419-432.
[9]. Kia M. Soft Computing in MATLAB.Qian academic publishing. [Persian]
[11].Demuth H,Beale M. Neural network toolbox for use with MATLAB. Sixth printing Revised for Version 4. Pp:680.
[12].Hahangeer AR, Raeini M, Ahmadi MZ. Comparison of artificial neural networks (ANN) simulation of rainfall-runoff process with HEC-HMS model in Kardeh watershed. Journal of Water and Soil. 2008. 22(2):72-84. [Persian]
[13]. Kumar S, Merwade V, Kam J, Thurner K. Streamflow trends in Indiana: effects of long term persistence, precipitation and subsurface drains. Journal of Hydrology. 2009.374(1): 171-183.
[14].Cybenko G. Approximation by superposition of a sigmoidal function. Mathematics of control, signals and systems 2.4. 1989. 303-314.
[15].Hornik K, Stinchcombe M, White H. Multilayer feed-forward networks are universal approximators. Neural Networks. 1989. 2(5):359-366.
[16].Zhang G, Patuwo BE, Hu MY. Forecasting with artificial neural networks: the state of the art. International Journalof Forecasting. 1998. 14(1):35-62.
[17].Noori R, Abdoli MA, Ghasrodashti AA, JaliliGhazizade M. Prediction of municipal solid waste generation with combination of support vector machine and principal component analysis: a case study of Mashhad.Environmental Progress & Sustainable Energy. 2009. 28 (2):249-258.
[18].Nikmanesh MR. Prediction of Monthly Average Discharge Using the Hybrid Model of Artificial Neural Network and Wavelet Transforms (Case Study: Kor River Pol-e-Khan Station). Journal of Water and Soil Conservation. 2015. 22(3):231-239. [Persian]
[19].Noori R, Farokhnia A, Morid S, RiahiMadvar H. Effect of Input Variables Preprocessing in Artificial Neural Network on Monthly Flow Prediction by PCA and Wavelet Transformation. Journal of Water & Wastewater. 2008. 20(69):1-22
Volume 3, Issue 4
January 2017
Pages 659-667
  • Receive Date: 21 November 2016
  • Revise Date: 27 December 2016
  • Accept Date: 30 December 2016
  • First Publish Date: 30 December 2016
  • Publish Date: 21 December 2016