| This thesis presents numerical simulations of tip vortex flow over a finite-span hydrofoil and a marine-propulsor. The study is focused on the tip vortex formation and the tip vortex flow in the near wake. The computational approach involves using a high-order finite-difference method to solve three-dimensional, incompressible Reynolds-Averaged Navier-Stokes equations.; Grid designs are based on the objectives of the study and the preliminary calculations. In order to study grid dependence, three grids are generated for each problem. For each grid, the iterative uncertainty is assessed. The verification analysis is performed through integral and point variables.; Four turbulence models are evaluated in the hydrofoil problem. They are the Baldwin-Lomax model, the k-ω baseline and shear stress transport models, and the Gatski-Speziale algebraic stress model. The numerical results correctly show the characteristics of the tip vortex formation. The spanwise pressure gradient is found over the wingtip. Flow is wrapped from the tip boundary layer into the tip vortex, which cause the streamlines converge and diverge on the suction side of surface. A strong shear layer with high turbulence is found near the tip. High gradients of pressure and velocities in the tip vortex are captured. Both k-ω turbulence models show good mean flow agreement with the data over the wingtip and in the near wake. Of the four models, the non-linear model gives the best numerical results. It predicts an isotropic turbulence in the tip vortex, which is consistent with the data.; Based on the study of the hydrofoil tip vortex flow, two turbulence models, the k-ω baseline and the Gatski-Speziale algebraic stress models, are used in the propulsor problem. Results from both models provide relatively good predictions of propulsor performance and mean flow, with the exception of the flow in the hub boundary layer. The tip vortex flow is simulated well. The predicted minimum pressure in the tip vortex agrees with cavitation inception, within the range of experimental uncertainty. Similar to the hydrofoil problem, the non-linear model provides better resolution of tip vortex flow. |
