Flow3D-CFD for predicting sediment scouring and deposition due to Pump Storage Hydropower operations

With the increasing dependance on renewable and infrequent energy sources, such as wind and solar energy, the need for long-term storage solutions such as Pump Storage Hydropower (PSH) is becoming more necessary to balance supply and demand of energy all-year round [1]. Sediment dynamics is a crucial factor in the lifetime assessment of powerplant operation and efficiency; sediment scouring and deposition are significant processes that might be triggered by hydropower operations. These negatively impact the powerplant in many aspects: e.g., channels section change, reservoir siltation, hydraulic machinery erosion [2]. Dredging, flushing operations and venting of turbidity currents are common measures used to remove or mitigate further depositions. Numerical models are the most common tool in judging hydropower operation and the resulting effect on intake and outlet structures in terms of hydrodynamics and sediment transport parameters. In cases where the focus is at the scale of river channel section and related hydraulic works, the use of 3D Computational Fluid Dynamic models provides the necessary details to understand the hydrodynamics and sediment behavior. This study aims to investigate the resulting effects of PSH operation at inflow/outflow typical structures in terms of erosion and deposition. This generates a submerged jet characterized by mean flow strong momentum and high turbulent energy content. The pipe outflowing jet exerts shear stress at the channels loose bed, resulting in scouring. In addition, shear layer develops between inflowing jet and steady water in the basin which might promote the advection and diffusion of eroded sediment eventually settling further downstream. The model was developed on FLOW3D, utilizing both Reynolds Averaged Navier Stokes (RANS) and Large Eddy Simulation (LES) turbulence models [3]. A packed sediment layer has been applied to the domain, consisting of sand and gravel. Preliminary results show that RANS produces a robust prediction of mean velocity fields where the higher values are concentrated along the jet axis. LES yields more variable water velocities and sediment suspension and erosion fields; this is important when investigating shear layer dominated conditions such as local scouring/deposition and re-suspension. Further modifications to the model are planned, such as adding unsteady and oscillating inflow and refining sediment gradation to better represent possible prototype cases.