Evaluation of Flow Resistance and Estimation of Bed Load Transport in Gravel-Bed Channels

Abstract

A reliable estimation of roughness coefficient and Bed Load transport rates is necessary for the hydraulic analysis of open channels and evaluation of change in conveyance in a specific flow condition. Determining flow resistance and transport parameter in Bed Load transport conditions is difficult because they depend upon the strength of the flow, the complex structures of secondary flow, bed shear stress, and bed particle sizes. For gravel-bed streams, many researchers have analysed flow conditions over movable channel beds and proposed equations for Darcy-Weisbach friction factor (f) and Einstein Bed Load transport parameter (p) for limited geometrical and hydraulic conditions. Movable Bed Load conditions cause many variations of the influencing parameters, which complicate the determination of flow resistance by employing analytical methods. The gravel-bed channel response for a movable Bed Load condition in terms of resistance to the flow is distinct from that of a fixed bed and requires a different technique evaluation. The present study considers the different empirical approaches by various authors to calculate the friction factor and Bed Load transport parameter under movable bed load conditions and propose expressions using advanced computing methods, viz. Artificial Neural Network (ANN) and Genetic Expression Programming (GEP). Various hydraulic and geometric parameters affect flow resistance and transport of Bed Load. A wide range of experimental datasets is used to investigate the effect of these influencing parameters. In the present study, the factors that influence friction factor and Bed Load transport parameter for such a flow condition are the relative submergence depth, bed slope, aspect ratio, Reynolds number, Froude number and shield number. New models have been developed to estimate the roughness coefficient and Bed Load transport rates. The predictability of the proposed model is compared to the various empirical equations in the literature. Unlike the existing models, the proposed models are observed to effectively predict the friction factor and Bed Load transport parameter for many datasets. The evaluated value of the friction factor from different models is used to validate the conveyance capacity of a river. The developed Multi-Gene Genetic Programming (MGGP) model is also observed to reasonably predict discharge in the river, signifying that the model is competent to be applied to field conditions within the specified range of parameters