VON KARMAN VORTEX SHEDDING STUDY WITH FOAM-EXTEND 4.0

When exploited over their full range of operation, hydraulic turbines experience several forms of fluid-structure interactions (FSI) phenomena. Among these FSI phenomena, periodic shedding of vortices in blade wakes, known as von Karman vortex shedding, constitutes one of the most prominent unsteady flow behavior, which may lead to noise, vibration and fatigue damage. Vortex shedding frequency and wake pressure variation amplitude must therefore be predicted accurately at the machine design stage. Such predictions are however difficult and expensive to perform numerically because of several flow features which simultaneously contribute to render simulations hard to perform. Indeed, at typical model scale, experimental investigations have shown that the accelerated fluid over the profile experiences a transition between laminar and turbulent flow. Boundary layer instability at the profile trailing edge constitutes the root mechanism that initiates vortex shedding. The shedding frequency is therefore highly dependent on the accurate prediction of the boundary layer thickness and evolution over the profile, thereby requiring advanced transitional turbulent flow modeling to accurately capture the boundary layer behavior. Once shed in the profile wake, vortices are transported in the core flow and must be allowed to diffuse prior to reaching the computational domain outlet, to insure computational stability. The computational domain must therefore be extended by several profile chord lengths to prevent vortices from interacting strongly with the outlet boundary, thereby significantly increasing simulation computational cost. The present article will investigate unsteady CFD simulations of von Karman vortex shedding using Foam-extend-4.0 Simulations will be based on the k-omega SST turbulence model, and the inclusion of a two equation γ-Reθ transitional turbulence model will be examined. Simulation results will be compared to available experimental data on a truncated NACA 0009 hydrofoil [1,2,3], and with results computed using a commercial code. Two mesh generation approaches will also be compared; first, a hybrid approach and second, a multi-block structured approach, including hexahedral meshes.

Results of pressure and velocity fields will be analysed. Monitoring points added in the wake of the hydraulic profile will be used to determine the vortex shedding frequency, which will be calculated using Fast Fourier Transformation (FFT) of the simulated signals. Detailed comparisons of both frequency and amplitude will be carried out.

References [1] P. Ausoni, “Turbulent vortex shedding from a blunt trailing edge hydrofoil,” PhD Thesis, École Polytechnique Fédérale de Lausanne, Switzerland, 2009.

[2] T. Vu, B. Nennemann, P. Ausoni, M. Farhat and F. Avellan, "Unsteady CFD Prediction of von Kármán Vortex Shedding in Hydraulic Turbine Stay Vanes," Hydro 2007, Granada, Spain, 2007.

[3] A. Zobeiri, “Effect of Hydrofoil trailing edge geometry on the wake dynamics,” PhD Thesis, École Polytechnique Fédérale de Lausanne, Switzerland, 2012.