Wind Energy Applications of Unified and Dynamic Turbulence Models

This study investigates linear and nonlinear dynamic LES and unified RANS-LES models obtained by stochastic analysis regarding the simulation of the atmospheric boundary layer. The advantages of dynamic LES methods are their accuracy, the ability to account for anisotropy and backscatter, and their computational stability. The advantage of unified RANS-LES models is their ability to provide results like LES for a fraction of the cost of LES. Two applications are considered: (i) turbulent boundary layer simulations on fine grids, which were performed to demonstrate the suitability of models for turbulent boundary-layer turbulence studies, and (ii) neutral boundary layer simulations on coarse grids, where subgrid models have to contribute significantly. On fine grids, all models perform comparably. On coarse grids, a nonlinear dynamic model provides the most accurate results. A nonlinear unified RANS-LES model is shown to be well applicable as a low cost alternative.