Unsteady atmospheric flow solutions with OpenFOAM coupled with the numerical weather prediction software, WRF

---

Tezin Türü: Doktora

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Havacılık ve Uzay Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2018

Öğrenci: ENGİN LEBLEBİCİ

Danışman: İSMAİL HAKKI TUNCER

Özet:

Ways to predict wind energy potential and energy generation accurately are sought in many applications such as the wind turbine siting and the short term wind power production estimation. Current wind energy prediction models are based on the statistical analysis of long term observation data and the reconstruction of average wind fields by means of numerical tools. Even if computational fluid dynamics tools based on Navier-Stokes solutions are employed for sectoral wind fields, the reconstruction of wind fields according to prevailing wind speeds and directions are, in general, not accurate for wind field predictions on complex terrains.On the other hand, numerical weather prediction software such as WRF, in which the unsteady atmospheric physics are taken into account, provides time dependent atmospheric flow field solutions, but due to the low resolution and the pressure based vertical coordinate system wind field predictions are inaccurate near ground in complex terrains. In this work, unsteady atmospheric flows are computed by a coupled solution methodology. Atmospheric flow solutions based on the weather prediction model, WRF, on a low resolution grid provide the unsteady boundary conditions for OpenFOAM solutions on high resolution terrain fitted grids. The usage of WRF predictions as boundary conditions accounts partially for the physical processes such as solar radiation, precipitation, surface heating and the resulting diurnal cycle. OpenFOAM solutions with the high resolution topography and computational grid, and a spatially varying roughness model capture complex terrain effects such as flow separation, recirculation, and reattachment zones. The OpenFOAM solutions coupled with WRF are performed in parallel. A post-processing tool is also developed to extract the wind field data from the coupled OpenFOAM solutions at any given altitude, and to estimate the distributions of wind power density and wind energy generation by time integration along the unsteady flow solution. The coupled flow solutions are compared with the time series wind speed and direction data obtained from a meteorological measurement tower, Met-mast. It is shown that the coupled flow solutions developed improve the time dependent wind field predictions of WRF by about 12% with respect to the observation data which corresponds to about 40% improvement on wind power production estimations.