Sediment transport and bed evolution in a 180∘curved channel with lateral intake: Numerical simulations using Eulerian and Discrete Phase models

Abstract

Lateral intakes are hydraulic structures used for domestic, agricultural and industrial water conveyance, characterized by a very complex three-dimensional morphodynamic behavior: since streamlines near the lateral intake are deflected, some vortices form, pressure gradient, shear and centrifugal forces at the intake generate flow separation and a secondary movement, responsible for local scour and sediment deposition. On the other side, the modeling of flows, besides the sediment transport, in curved channels implies some more complications in comparison with straight channels. In this research, this complex process has been investigated experimentally and numerically, with the mechanism of sediment transport, bed topography evolution, flow pattern and their interactions. Experiments were performed in the Laboratory of Tarbiat Modares University, Iran, where a U-shaped channel with a lateral intake was installed and dry sediment was injected at constant rate into a steady flow. Due to the spiral flow, the bed topography changes significantly and the bed forms in turn affect the sediment entering the intake. Different from the previous works on this topic which were mainly based on laboratory experiments, here, Computational Fluid Dynamics (CFD) numerical simulations with FLUENT software were also performed, specifically with the two-phase Eulerian Model (EM) and Discrete Phase Model (DPM), at the aim of evaluating their performance in reproducing the observed physical processes. This software is used for a large variety of CFD problems, but not much for simulating sediment transport phenomena and bed topography evolution. The comparison of the results obtained through the two models against the laboratory experimental data proved a good performance of both the models in reproducing the main features of the flow, for example, the longitudinal and vertical streamlines and the mechanism of particles movement. However, the EM reveals a better performance than DPM in the prediction of the secondary flows and, consequently, of the bed topography evolution, whereas the DPM well depicts the particles pattern, predicts the location of trapped particles and determines the percentage of sediment entering the intake.The numerical models so calibrated and validated were applied to other cases with different positions of the intake in the bend. The results show that mechanism of sediment entrance into the intake varies in different position. If the intake is installed in the second half of the bend, the sediment accumulates along the inner bank of the bend and enters the intake from downstream edge of intake; on the other side, if it is placed in the first half of the bend, the sediment accumulates along both the inner and the outer bends and, therefore, more sediment enters the intake. Also the results of the simulations performed with the DPM model for different positions of the lateral intake show that for all discharge ratios, the position of 120∘is the one which guarantees the minimum ratio of sediment diverted to the intake (Gr).