D simulation of flow pattern in the vicinity of the inlet canal of lateral intake diversion dam

Direct outlets from rivers are the most common structures to supply water for agriculture, industry and drinking water uses. Among the various methods of river intake, construction of a diversion dam and lateral intake is the most common way to supply water for irrigation network. This method is based on diverting a part of the river flow using a diversion dam. Since the river flows often contain sediment load, sedimentation usually occurs near the intake structure. Consequently, during the time it causes blocking the intake inlet and decreasing capacity of intake canal. Many researches have been done about the optimal design for connection between a main canal and the intake inlet and also different methods have been suggested to control sedimentation. Often they have not considered the diversion dam in the vicinity of intake inlet. However, the comprehending of flow and sedimentation patterns requires the attention to all components of a flow diversion system. Nowadays, with progression of numerical models, the application of numerical simulation of flow pattern in such structures is provided. Among numerous released software for numerical simulation, the Fluent is a powerful software for analysis and simulation of fluid flow which widely used by researchers. In the Fluent the users can define new physical properties and boundary condition with especial functions. This software uses a finite-volume method for solving the Reynolds averaged Navier- Stokes equations. In this research, three-dimensional simulation of flow pattern in the lateral intake, with taking into account diversion dam structure, dividing wall, sluice way and entrance sill height was done by using the Fluent. The results of numerical model are validated with experimental tests which had been carried out in the Irrigation and Reclamation laboratory of Tehran University. Then different characteristics of the flow separation zone, flow patterns, vortex, and secondary flows were compared. The experimental tests consist of a Plexiglas flume with a lateral canal, main canal, diversion dam structure, dividing wall, and sluice way. The main canal consists of a rectangular cross-section with a width of 0.90 m, height 0.60m and 18m in length. The lateral intake with 0.4 m in length and width was installed at 120° (at the left side) to the main canal. The longitudinal slope of the flume was set to 0.0008. The diversion canal and sluice way were perpendicular to flow alignment of the main canal. The width and height of diversion dam are 0.6 and 0.27 m, respectively. The lateral intake was separated of diversion dam with the sluice way. It is 0.29 m in length and 0.1 m width. In the first, the canal invert was covered with sediment, and then flow characteristics were measured at different levels and the diversion ratios of 0.5 and 0.65. In this study, the magnitude of flow velocity (z velocity) near the intake and sluice way is used for verification of numerical simulation. The 3D geometries and meshes of intake system were generated by using the Meshing Tool of Gambit software. Computational domain was built with 218000 grid nodes. In this study, the governing equations are discretized by first order upwind method and the pressure is decomposed from the velocity by SIMPLE algorithm. The standard k-ε turbulent model is used to describe the flow separation. The Volume of Fluid (VOF) method is used to track the fluid interfaces. The considered boundary and initial conditions included: (1) inlet velocity condition and a water depth of 23 centimeters is assumed at the main canal inlet, (2) the outflow condition is used at the intake outlet, and (3) it is assumed that the sluice way is closed. The results of numerical model showed that the flow pattern in the vicinity of the intake canal has been simulated good. Comparison of small amounts of z velocity between experimental and numerical models, particularly in diversion ratio equal to 0.5, indicated suitable model reliability for simulation in the lateral intake. The result showed that the average error rate of the numerical model at three different levels was about 13%. After model verification, impacts of shape of the entrance intake on the flow pattern formed at the intake inlet have been studied. The results of the numerical model for angled upstream wall inlet, revealed that the low-velocity zone was removed from this region, as the width of the eddy zone has been reduced from about %8 to near zero.