Abstract:
The problem of transformation of rainfall into runoff has been a subject of scientific investigations throughout the evolution of the subject of hydrology. A number of investigators have tried to relate runoff with the different characteristics which affect it. Various researchers have attempted to address this modelling issue either using knowledge based models or data-driven models. However, simulating the real-world relationships using these Rainfall-Runoff models is not a simple task since the various hydrological processes that involve the transformation of rainfall into discharge are complex and variable. In recent years, data-driven soft computing techniques e.g. artificial neural network and fuzzy logic have gained significant attention in hydrological modelling. In the present paper fuzzy rule based approach is chosen for developing a rainfall-runoff model for Manot sub-basin of Narmada River system. Further, the model performance has been examined using global model performances indices. The results of the study indicate that the choice of the model input structure certainly has an impact on the model prediction accuracy. The fuzzy model has improved with the increase in the number of input combinations up to a certain extent. The study presents an efficient methodology developed for rainfall runoff modeling over the medium size catchment with limited data.
