Experimental and numerical studies for estimating coefficient of discharge of side compound weir

Abstract

A sharp-crested side compound weir is a flow diversion structure provided on one or both side walls of a channel to divert water from the main channel. Compound sharp-crested weirs are widely used in irrigation, hydraulics, and environmental engineering. This article presents results of experimental and numerical studies conducted on sharp-crested side compound weirs in open channels. Owing to the complex mechanism of flow through a side compound weir it is difficult to establish a regression model to accurately predict the coefficient of discharge (Cd). In this study, an alternative approach to the conventional regression modelling in the form of artificial neural network (ANN) has been used to predict the values of Cd. A network architecture with trained values of connection weights and biases is recommended to predict Cd. The input to ANN model consists of grouped parameters pertaining to the ratio of weighted crest height to the length of the side compound weir ([Formula: see text]), the ratio of upstream depth to length of the side compound weir (Y1/L), and upstream Froude number (F1). The results of the ANN model applied herein were found to be superior to those obtained through regression modelling by previous researchers. The sensitivity analysis of the ANN model shows that [Formula: see text] is the most important parameter for the estimation of Cd; followed by Y1/L and F1.