Abstract

A recently developed non-hydrostatic three-dimensional RANS model to predict flow and sediment transport in curved open channels (Zeng et al., 2006) is tested for the case of the flow in a channel with very strong curvature corresponding to the experiment of Blankaert (2002). The model uses fine meshes that do not necessitate the use of wall functions, a deformable free surface model, and a linear Spalart-Allmaras turbulence model that can account for the effects of bed roughness. The bed is flat corresponding to the start of the erosion process. It is shown the model successfully captures the distribution of the streamwise velocity component and the free surface profile at representative cross sections along the curved bend. Though the model captures the main circulation cell associated with the classical helical motion, it does not capture the weaker circulation cell present at the outer bank in experiment. Based on the model predictions, the distributions of the streamwise vorticity, turbulent kinetic energy and the bed shear stress are analyzed.

Keywords: Open channel flows, Turbulence modeling, Strongly curved bends