Installation of a tidal turbine at a site requires detailed information of the bathymetry and the surrounding flow field for its deployment and efficient operation. Increasingly, Computational Fluid Dynamics (CFD) is being used to predict the local unsteady flow field. Bathymetry data is generally of a different resolution than required for CFD meshing, and requires special treatment for incorporation into the meshing process. This process results in resolved and unresolved bathymetry features that may be important in obtaining accurate CFD predictions. This issue is explored at a tidal site located in Minas Passage which connects the Minas Basin to the Bay of Fundy. Bathymetry processing is considered in conjunction with a high resolution Detached Eddy Simulation (DES) of turbulent tidal flow obtained using a 26 million node mesh. This particular site has a steep ridge which results in significantly large unresolved data relative to the mesh size, specifically at the ridge. A methodology is suggested to separate the unresolved data into sub grid and supra grid scales. The sub grid data is modelled using the sand grain roughness model while supra grid data is modelled using the proposed form drag model. Predicted results are compared to Acoustic Doppler Current Profiler (ADCP) measurements for one tidal site under ebb conditions for an inlet velocity of 4 m/s. The improved model gives results closer to the experimental data as compared to the sand grain roughness model.
Zircon - This is a contributing Drupal Theme Design by WeebPal.
