Understanding flows over channel floor with cross-sectional joints

Abstract

One of the main elements of hydraulic plants are the spillways. The main purpose of this element is to discharge water from the basin when it reaches levels above the safety limits. Therefore, spillways are designed to withstand large water flows and high pressures. However, poor channel design can lead to high pressures at the joints of the concrete slabs that make up the channel and result in fatal landslides. Clear examples are the accidents at Oroville Dam in 2017, Dickinson Dam in 1954, and Big Sandy Dam in 1983. The objective of this project is to develop a numerical model to analyze the behavior of the flow over the spillway and to study the relationship between the flow velocity and the uplift pressures generated at the channel slab joints. For this purpose, a lot of research have been made in the field trying to understand the water behavior over a spillway, but few have focused on the uplift pressures under the slabs due to the stagnation of the water. Some experiments made by the Bureau of Reclamation analyzed the flow over the joint for different situations. In 1976, Mr Perry L. Jonshon studied experimentally the pressures generated on the linings in an open channel, while Warren Frizell tested, in 2007, the pressures under the slabs in a close water tunnel. The aim of the current project was to develop a numerical model to predict accurately the behavior of the flow over the joints and compare it to the experimental results. Computational Fluid Dynamics (CFD) has been used as the main tool in the numerical analysis. This software has been continuously used by many scientists during the years and many investigations have verify the accuracy of the tool to predict the fluid over a spillway. Three different joint geometries have been studied, sharp-edged, chamfer-edged and radius edged joints. All the cases have been analyzed for sealed and vented situations and compared between them. The current project has obtained great results regarding the uplift pressures and flow behavior over the joints compared to the experimental tests. The numerical model shows a maximum of 6% of deviation related to the experimental results for sealed configurations, while the vented configurations take deviations of around 15%. This increase is expected to be due to the lack of information of the discharge flow rate, for this reason, the results related to the analysis over a slope with a free surface are considered trusty enough. The study shows similar results for different slopes configurations which explains that the gravity is not the main force of the study as it was previously expected