Author(s): Shreenivas N. Londhe; Kirti P. Sonawane
Linked Author(s):
Keywords: Stage-discharge; Artificial neural network; Radial basis function; Feed forward back propagation
Abstract: Reliable prediction of discharge in a river is vital component for hydrological and hydraulic analysis. A graph of number of concurrent observation of stage vs. discharge over a period of time represents stage-discharge relationship, also known as rating curve. Once a relationship is established, it can be used to transform the observed stage into corresponding discharge. Forecasting of stage-discharge is of enormous importance for reliable planning, design and management of water resource project. Data driven technique like ANN is a powerful tool for input-output mapping and can be effectively used for reservoir inflow forecasting and operation. In this paper, models of stage-discharge relationship are built up using two different Artificial Neural Networks such as Feed Forward Back Propagation (FFBP) and Radial Basis Function (RBF). The daily data at two stations in Krishna river basin, India having stage and discharge values are used to estimate discharge when stage is used as input. The results of both the networks are compared with each other.
Year: 2018