Wednesday, 9 January 2013
Exhibit Hall 3 (Austin Convention Center)
Our study documents the improvement offered to wind-farm forecasts by WRF's large-eddy simulation (LES) when such simulations' initial and boundary conditions were produced by WRF's wind-farm drag parameterization scheme (WFDPS). With version 3.3, the WRF model included WFDPS that represents sub-grid effects of specified turbines on wind and turbulence kinetic energy (TKE) fields. The study examines the effect of using the WFDPS's parameterized TKE solution to initialize LES's more explicit turbulence model. The Cedar Creek Wind Farm in northeast Colorado is the site in question. Statistical analysis is performed, comparing the output from the MYNN planetary boundary layer scheme without WFDPS, MYNN with WFDPS, 3d TKE LES scheme initiated by WFDPS, and 3d Deformation (Smagorinsky) LES scheme initiated by WFDPS. Performance of each LES scheme is examined across the range of its respective sub-grid coefficient. The models' performance at downwind turbines receives particular attention. The study examines 30 cases, including three periods that were executed without WFDPS and published by NCAR. Our model results are compared to NCAR's published results.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner