ANFIS-based approach to predicting scour location of spillway

Abstract

A spillway with a trajectory bucket as an energy dissipator satisfies the requirements of both safety and economy. Many researchers have developed empirical formulae to predict scour location considering different hydraulic parameters and bucket configurations. An attempt has been made in this paper to estimate the location of maximum scour using parameters such as q, H1, R, d50, dw and the lip angle of the bucket using an adaptive neuro fuzzy inference system (ANFIS). Extensive experimental results were collected and analysed in order to investigate the effect of different parameters on throw distance and scour location downstream of a trajectory bucket spillway. It was found that the scour parameter is a function of the fall Froude number, ratio of head to tailwater depth, ratio of radius of trajectory bucket to head, ratio of sediment mean size to head and lip angle of the bucket. The functional relationship is expressed in dimensionless form. This paper also presents an alternative to the regression in the form of neural fuzzy modelling. The results of this modelling were compared with the regression equation and it was found that the ANFIS results are highly satisfactory. The results of this study can be used to predict the location of maximum scour downstream of the trajectory bucket spillway.