A Modified Coherence Method for Flow Prediction in a Compound Channel

Abstract

The present thesis is a result of physical and numerical research undertaken by the author to understand the flow characteristics like depth-averaged velocity distribution, boundary shear stress distribution, discharge, etc. of an asymmetric compound channel with differential roughness. Two sets of experiments have been conducted on an asymmetric compound channel in which both the main channel and flood plain are of trapezoidal cross-section. For the first set of experiments, the main channel was kept smooth whereas the flood plain was made rough by laying plastic mat on its bed to achieve differential roughness. The second set of experiments have been conducted on the same asymmetric compound channel but with different roughness condition. The main channel was kept as it is, but the roughness of flood plain again changed by fixing small gravels on it. Using micro-ADV, three dimensional velocity data have been collected from the main channel flow. To get the flow velocity of flood plain flow, Pitot was used. Boundary shear stress values were also measured using Preston tube technique for the compound channel.

Depth-averaged velocity distributions and boundary shear stress distributions for all flow depths of two sets of experiments for the whole asymmetric compound channel section were plotted and analysed. Energy correction factors and momentum correction factors for all flow depths of the two sets of experiments were calculated. Weighted divided channel method (WDCM), which was developed by Lambert and Myers (1998) was applied to both set of experiments and the values of weighting factors for flood plain and main channel were calculated. A wide range of data sets, which includes symmetrical and asymmetrical compound channels with homogeneous as well as differential roughness, were considered in this study to check the accuracy of Coherence method, which was first developed by Ackers (1991). A new model has been developed to evaluate discharge adjustment factor (DISADF), from which discharge can be predicted easily. The new is found to be capable of predicting the discharge with satisfactory results by resulting an average error of 4.6248% for symmetrical compound channel data sets considered for this study and 4.0927% for all the asymmetrical compound channels. The prediction capacity of the model has been validated with experimental data sets of other researchers, FCF data sets and natural river data sets.