A spillway is a critical hydraulic structure designed to regulate water level, prevent overflow and ensure discharge from the reservoir to downstream. A spillway can experience cavitation if the water velocity exceeds 20 m/s, requiring its design to be optimized. Many existing spillway devices were constructed with lower capacities and are inadequate to discharge the current design water flows. Cavitation in spillways with high head poses a risk to structural integrity and dam safety. Aeration devices mitigate bubble collapse and subatmospheric pressure conditions, which reduces the risk of cavitation damage to concrete surfaces. The investigation of such flow behavior is complex due to the high turbulence and unsteady nature of the two-phase flows.
To evaluate such scenarios, laboratory models and physical tests are often used to investigate the load-carrying capacity and energy conversion. However, accurately measuring airflows under high flow conditions remains challenging. Numerical calculations (CFD) complement physical experiments and enable independent hydraulic investigations, providing valuable insights into the behavior of mixed-phase flows.
This paper presents numerical studies of two-phase flows and energy conversion in a spillway, focusing on high-velocity flow conditions that are critical for dam safety. The main objective is to use numerical tools to assess the hydraulic performance of the spillway and to provide practical guidance for design and operation. Along with a hierarchy of turbulence models, including Reynolds-Averaged Navier-Stokes (RANS), Detached Eddy Simulation (DES), Delayed Detached Eddy Simulation (DDES) and Large Eddy Simulation (LES), the study uses advanced CFD techniques, including the so-called Volume of Fluid (VOF) model and two-phase models. The numerical models are validated against experimental data.
Paper I presents an investigation of four geometric designs of flat and pooled step-like discharges. The main results from Paper I show that pooled step configurations improve energy conversion and reduce the risk of cavitation compared to flat step designs. In addition, the pooled steps exhibited superior overall hydraulic performance and improved flow behavior, by achieving the lowest downflow velocities and effectively minimizing the cavitation potential. Furthermore, Paper II investigates the comparative performance of 2D and 3D modeling methods, together with advanced turbulence models, for simulating two-phase flow of air and water. The results indicate that 2D Eulerian models are suitable for shallow water flows with negligible vertical velocity components and are appropriate for preliminary analyses due to their computational efficiency and reasonable accuracy. In contrast, 3D models are better suited for highly accelerated and highly turbulent flows, and offer improved accuracy, especially in representing complex flow characteristics. Advanced turbulence resolving methods, such as DES and DDES, provide improved resolution of the flow field and more reliable predictions of air entrainment. In addition, the application of 3D turbulence modeling improves the description of air-water flow behavior and increases the accuracy of the prediction of air concentration in the impact zone.
Next, Article III examines the sensitivity of aerated discharge predictions to grid resolution using Sub Grid Scale modeling (SGS). The results show that grid resolution with Locally Refined Structured Meshes (LRSM) exhibits improved accuracy in capturing complex air-water interactions compared to Unstructured Meshes (USM). Furthermore, aerated discharge predictions are significantly more sensitive to grid resolution than to the choice of SGS modeling approach.