Modelling runoff in a river basin, India: an integration for developing un-gauged catchment. (7th May 2020)

• Record Type: Journal Article
• Authors: Samantaray, Sandeep; Ghose, Dillip K.
• Is Part Of: International journal of hydrology science and technology. Volume 10:Number 3(2020)
• Page Start: 248
• Page End: 266
• Publication Date: 2020-05-07
• Subjects: artificial neural network -- ANN -- regression -- river basin -- runoff -- stage -- India; Hydrology -- Periodicals; Water resources development -- Periodicals; Hydrology; 553.705
• Languages: English
• ISSNs: 2042-7808

Abstract

Stage-runoff model based on nonlinear multilayer regression (NLMR) and artificial neural networks (ANNs) are developed in the present study. Models are developed using collected dataset in short term basis during monsoon. The results confirmed that back propagation neural network (BPNN) model is an important alternative to regression models. BPNN are developed using extended gradient descent-based delta-learning algorithm and radial basis function network (RBFN) are developed using Gaussian potential functions. Predicted results using BPNN and RBFN model perform better as compared to NLMR and BPNN is found to be the best among all three techniques. The results of this work are integration for measuring runoff in un-gauged catchment approaching to the river basin.