MODELLING OF MONTHLY RAINFALL-RUNOFF PROCESS USING ANN

Abstract

Prediction of hydrological variables, like rainfall, runoff, river flows etc. is necessary in planning design, maintenance and operation of water resources systems. The process of rainfall-runoff (R-R) is highly nonlinear, time varying, spatially distributed, and therefore cannot be easily described by simple mathematical models. An Artificial Neural Network (ANN) is a flexible mathematical structure that is capable of identifying complex nonlinear relationships between input and output data sets. ANN models have been found useful and efficient, particularly in problems for which the characteristics of the processes are difficult to describe using physical equations. Therefore, the present study is attempted to propose a modeling approach that couples a system theoretic model with the ANN for estimation of monthly runoff. Realistic results are obtained through non-updating monthly rainfall-runoff modeling for the small catchments of a few large drainage basins by using the auxiliary model output as one of the inputs for ANN based modeling.