Record Details

Lumped and distributed data ANN models for an intermittent river

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Lumped and distributed data ANN models for an intermittent river
 
Creator MAGAR, R.B
SUNIL, K
JOTHIPRAKASH, V
 
Subject Hydrologic models, Rainfall-runoff relationship, Artificial neural network, Lumped data, Distributed data, Intermittent river.
 
Description This study presents the application of Artificial Neural Network (ANN) models for intermittent rainfall-runoff processes using lumped and distributed data. The developed Multilayer Perceptron (MLP) ANN models are trained with Back propagation (BP), Conjugate Gradient (CG), and Lavenberg-Marquardt (LM) algorithm. All the networks are trained with various combinations of length of data, structure, activation function, momentum factor, learning rate and number of epochs. Since the rainfall-runoff cross correlation is good, both lumped and distributed data ANN models performed equally better. However, from the cause effective ANN models it has been found that the BP algorithm performed well for lumped data and CG algorithm performed well for distributed data. It has also been found that the distributed model performed better in mapping the rainfall-runoff relationship for an intermittent river system.
 
Publisher Serialpublication
 
Date 2012-07-17T10:24:34Z
2012-07-17T10:24:34Z
2010
 
Type Article
 
Identifier Journal of Flood Engineering,1(2)159-173
0976-6219
http://dspace.library.iitb.ac.in/jspui/handle/100/14387
 
Language English