The hybrid wind farm parameterization

Archer, Cristina L.The goals of this research are to improve the current treatment of wind farms in large-scale models via a new hybrid wind farm parameterization to better understand the potential impacts of offshore wind farms on the environment. ☐ Wind turbines extract energy from the atmosphere and the resulting wakes affect the boundary layer and the environment. The approach chosen in this dissertation to study these impacts of wind turbines is numerical simulations utilizing the weather prediction model called Weather Research and Forecast (WRF), in which wind farms are currently parameterized as simple elevated sinks of kinetic energy and sources of turbulent kinetic energy (TKE). A well-developed wind farm parameterization is essential to better understand the potential impacts of wind farms on weather and climate at the regional to global scales. ☐ The first part of this dissertation is therefore a case study to understand the impacts of hypothetical, large, offshore wind farms on local meteorology, especially the precipitation, employing the current two most widely used wind farm parameterizations. This study quantitatively tests whether the offshore turbines may affect precipitation patterns during Hurricane Harvey, since Hurricane Harvey brought to the Texas coast possibly the heaviest rain ever recorded in U.S. history, which then caused flooding at unprecedented levels. Model results indicate that the offshore wind farms have a strong impact on the distribution of accumulated precipitation, with an obvious decrease onshore, downstream of the wind farms, and an increase in the offshore areas, upstream of or within the wind farms. The accumulated precipitation during Harvey was reduced by up to 21% in the presence of offshore wind farms consisting of hundreds of thousands of turbines. Compared with the control case with no wind turbines, increased horizontal wind divergence and lower vertical velocity are found where the precipitation is reduced onshore, whereas increased horizontal wind convergence and higher vertical velocity occur upstream or within the offshore wind farms. The sensitivity to the size of the offshore array, the inter-turbine spacing, and the details of the wind-farm parameterization is assessed. ☐ In the second part of this dissertation, a new hybrid wind farm parameterization (a.k.a. hybrid model) is developed, which is not based on physical processes or conservation laws, but on a multiple linear regression of the results of sophisticated, high-resolution large-eddy simulations (LES) with simple geometric properties of the wind farm layout. The need for the new hybrid model arises from three previously-unknown weaknesses in the current wind farm parameterization in WRF, i.e., it neglects the effects of wind direction, it is insensitive to the relative position of the wind turbines within the farm, and it injects excessive TKE in the atmosphere. The new hybrid parameterization, however, successfully remedies these weaknesses. After validations against observations collected at an existing offshore wind farm (Lillgrund in Sweden) and against LES results at three hypothetical wind farms, the wind speed deficit and TKE predicted with the hybrid model are found to be in excellent agreement with the LES results and the wind power production estimated with the hybrid model also performs well compared with the observation data. In conclusion, wind turbine position, wind direction, and added TKE are essential to properly model wind farm effects on the surroundings and the hybrid wind farm parameterization is a promising tool to incorporate them in meso- and large-scale simulations.University of Delaware, School of Marine Science and PolicyPh.D