Rainfall-runoff modelling of Naula watershed using Adaptive Neuro-Fuzzy Inference System techniques

Abstract

The persistent sequence of rainfall-runoff data is an important hydrologic data for the nationwide or local flood management allocation plans and the determination of the storage capacity of the hydraulic facilities. An adaptive neuro-fuzzy inference system technique was used for rainfall-runoff modelling of Naula watershed in this study. Different combinations of rainfall and runoff lags were considered as the inputs to the model, and runoff of the current day was considered as the output. Input space partitioning for model structure identification was done by subtractive clustering approach. A hybrid learning algorithm consisting of back-propagation and least-squares estimation was used to train the model for runoff estimation. The optimal learning parameters were determined by trial and error using gaussian membership functions. Root mean square error and correlation coefficient were used for selecting the best performing model. The optimal learning parameters were determined by trial and error using gaussian type membership function for best fit cluster radius.