Comparison of effects of four subgrid-scale turbulence models in large eddy simulation of a large wind farm

The present study makes predictions of four different subgrid-scale (SGS) models in large eddy simulation (LES) of large wind farms. The considered models are Smagorinsky (SM), dynamic Smagorinsky (DS), dynamic Lagrangian (DL), and wall-adapting local eddy-viscosity (WALE). In all simulations, the atmospheric boundary layer (ABL) is neutral, and wind turbines are modeled using the actuator disk model (ADM). The impact of these subgrid-scale models on the large wind farm simulation is evaluated by studying the mean flow velocity, wake characteristics, wind turbine power extraction, and different mean kinetic energy budget terms. It is observed that various models show different power extraction, and WALE and DL models predict the maximum and minimum power extraction, respectively. Furthermore, the mean kinetic energy budget terms show discrepancies for different models. The dissipation term shows a significant discrepancy in the near-wall and turbine hub regions. This dissipation discrepancy is maximum between the Smagorinsky and dynamic models.