Description |
While Direct Numerical Simulations (DNS), Large Eddy Simulations (LES), Reynolds Averaged Navier-Stokes methods (RANS), and others have proven to be reliable methods to simulate flow through a wind farm, they are also computationally expensive. In this work, wind turbine parameterizations are implemented into the Quick Urban Industrial Complex wind model (QUIC-URB), a mass-consistent model used to produce time-averaged 3D wind fields, enabling QUIC-URB to model wind farm flows. Three wake parameterizations are introduced: the Jensen model, the 2D Jensen model, and the Bastankhah model. Results from all three configurations are compared against LES data, which is run for the same wind farm configurations. The Bastankhah model is shown to provide the best agreement with LES data, as it has the most accurate estimate of wake velocity deficit, and is able to provide a good approximation of wake shape. Additionally, it is confirmed that these QUIC-URB simulations can be run at a fraction of the time and computational cost required for the same simulation in LES. For this work, the QUIC-URB simulations are run on one core, and the run time for the present cases is on the order of tens of minutes. For the same configurations, the run time for LES is on the order of 1 month when running on 128 cores. |