Flow Resistance Study for Rough Meandering Channels

Abstract

Almost every river found in nature are meandering rivers. So to understand the behaviour of the river, it is very important to study the meandering channels and its flow. A uniform flow in an open channel flow is always opposed by the channel boundaries. This resistance to flow depends on many geometric and hydraulic parameters. Generally, in one-dimensional flow analysis, a proper value of roughness coefficient is used to calculate the carrying capacity of the channel. Normally the same value of roughness coefficient is taken for the whole surface and also for all flow depths. For the meandering channels, roughness coefficient varies mainly with the aspect ratio, flow depths, channel slope and sinuosity of the channel. An appropriate value of roughness coefficient is of utmost importance to correctly predict the discharge value from Manning’s equations. In the present study, an attempt has been made to develop a model for predicting roughness coefficient for meandering compound channels. Theoretical as well as numerical modelling has been used to develop such models. For theoretical modelling dimensional analysis has been used whereas artificial neural network and genetic programing has been used for numerical analysis. A set of experimentations has been carried out to observe the stage-discharge relationships for different non-homogenous bed roughness. Different parameters such as relative depth, aspect ratio, bed slope, sinuosity and relative bed roughness, were considered as contributing factors affecting roughness coefficient, which were individually analyzed by varying only one of the parameters while keeping the rest constant. The experimental data sets as well as data sets of various researchers have been used for the development of a generalised model. The developed model has been validated with the data sets of other researchers and it was found that the model purposed better result in comparison to the previously developed models.