Evaluation of effective Parameters in the estimation of Runoff in Alkabeer Aljanobee Catchment using Elman Neural Network


Abstract in English

Accurately modeling rainfall-runoff (R-R) transform remains a challenging task despite that a wide range of modeling, either knowledge-driven or data-driven. knowledge-driven models need a large amount of parameters, so it suffers from plenty numbers of parameters, for this reason the hydrologists start looking for a simple modeling methods, that need a few parameters such as data _driven methods, so The present study amis to use artificial neural network, which is one type of this methods for modeling the relationship between rainfall and runoff in Alkabeer Aljanonbee river catchment in Tartous City. Elman Neural Network is depended on for prediction of runoff by testing twenty four models have different architectures. So all models have been tested by using different numbers of neurons in the hidden layer, by using nntool book, which is available in the Matlab program. The results of the research verify that the model which has each of temperature, relative humidity, evaporation and rainfall in the input layer with time delay equal to three days (0:-3), in addation to preveous value of runoff (-1:-3), gives a best performance for used data with mean square error equal to 2.8*10^-5, and correlation coefficient 0.96. So it has been reached that Elman network technology gives a good results in modeling the relation rainfall_runoff So it could be a good alternative instead of traditional approaches.

References used

SOLAIMANI, k. Rainfall-runoff Prediction Based on Artificial Neural Network (A Case Study: Jarahi Watershed). ISSN United States. 2009, 856-865
DHAMGE, N. ؛KADU, M. ؛ATMAPOOJYA, S. Rainfall Runoff Modelling Studies Using Artificial Neural Network. International J.of Multidispl. Research & Advcs. in Engg. (IJMRAE) India, Vol. 4, No. I, 2012, 27-38
LAFDANI, E.؛ NIA, A.؛ PAHLAVANRAVI, A. ؛ AHMADI, A.؛ JAJARMIZADEH, M. Daily Rainfall-Runoff Prediction and Simulation Using ANN, ANFIS and Conceptual Hydrological MIKE11/NAM Models. International Journal of Engineering & Technology Sciences Iran. 2013, 32-50

Download