International Journal of Pure & Applied Bioscience (IJPAB)
Year : 2017, Volume : 5, Issue : 6
First page : (1471) Last page : (1481)
Article doi: http://dx.doi.org/10.18782/2320-7051.6078

Comparative Study of Multilayer Perceptron and Radial Basis Function Artificial Neural Networks for Rainfall-Runoff Modeling in a Watershed

Sanjarambam Nirupama Chanu* and Pravendra Kumar
Department of Soil & Water Conservation Engineering, College of Technology, G. B. Pant University of Agriculture and Technology, Pantnagar - 263145 (U. S. Nagar) Uttarakhand, (India)
*Corresponding Author E-mail: linthoich@gmail.com
Received: 17.11.2017  |  Revised: 22.12.2017   |  Accepted: 26.12.2017  

 ABSTRACT

This paper compares the performance of two artificial neural network (ANN)- multilayer perceptron (MLP) and radial basis function (RBF) for modeling daily rainfall-runoff in a Himalayan watershed called Bino watershed situated at Almora and Pauri Garhwal districts of Uttarakhand, India. The time series monsoon data of rainfall and runoff between 2000 and 2009 were used to train and test the models. The best input combination was selected by gamma test (GT) technique. The performance of both the MLP and RBF neural network models were comprehensively evaluated in terms of indices viz. correlation coefficient (r), root mean square error (RMSE) and coefficient of efficiency (CE). The results of the study indicate that the choice of the network type has certainly an impact on the prediction accuracy of model. Both models performed satisfactorily for runoff predictions; however, the MLP model outperformed the RBF model. The r, RMSE, CE and R2 values for the best MLP model during testing were determined to be 0.92, 0.96 (mm), 0.80 and 0.85, respectively. Results show that ANN models are useful tools for rainfall-runoff modeling the hydrologic response with good accuracy in the study watershed.

Key words: Artificial neural network, Multilayer perceptron, Radial basis function, Gamma test, Rainfall- Runoff Modelling

Full Text : PDF; Journal doi : http://dx.doi.org/10.18782

Cite this article: Chanu S.N. and Kumar P., Comparative Study of Multilayer Perceptron and Radial Basis Function Artificial Neural Networks for Rainfall-Runoff Modeling in a Watershed, Int. J. Pure App. Biosci.5(6): 1471-1481 (2017). doi: http://dx.doi.org/10.18782/2320-7051.6078