Runoff prediction from Gaula river using Heuristic approaches
Loading...
Date
2019-07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
Runoff is the most complex and important elements of hydrologic cycle which needs to be understood well and is to be predicted in a very efficient manner. Runoff prediction is very important for countries which are very much prone to floods in a short period of time. It can cause various famines and diseases if not controlled in a proper way. Considering these facts, a study has been carried out to assess the daily
monsoon runoff prediction from Gaula river, Kathgodam, Nainital, Uttarakhand, India. The daily monsoon meteorological data of 11 years (1st June, 2008 to 30th Sept, 2018) were collected from Gaula barrage located at Kathgodam, Nainital, Uttarakhand, India. In the present study, multilayer perceptron artificial neural network (MLP-ANN) and Wavelet based artificial neural network (WANN) techniques were used to predict the daily monsoon runoff. The daily data for monsoon period (1st June to 30th September) of years 2008-2015 and 2016-2018 were used to train and test the models respectively. The lags were decided on the basis of Minitab statistical approach and all the possible input combinations were put to back-propagation algorithm and tan sigmoid activation function for training and testing of models. The performance of the models was evaluated qualitatively by visual observations and quantitatively using various performance indices viz. RMSE, correlation coefficient, coefficient of efficiency and Willmott index. The WANN model performed better than the MLP-ANN model.
Description
Keywords
null