Wind shadows impact planning of large offshore wind farms

Sara C. Pryor, and Rebecca J. Barthelmie

Acknowledgment of support from the U.S. Department of Energy, Office of Science, MultiSector Dynamics, Earth and Environmental System Modeling (MSD), and Regional & Global Model Analysis (RGMA) program areas.

DOI:https://doi.org/10.1016/j.apenergy.2024.122755

Abstract

The technical potential offshore wind resource greatly exceeds current electricity use. However, areas suitable for bottom-mounted wind turbines close to large coastal demand centers are limited. Thus, an increasing number of offshore wind farms will operate in the wake (or wind shadow) of other wind farms. Maximizing system-wide electricity production and overall energy extraction from offshore wind, while being cognizant of other constraints, requires optimal siting of offshore wind energy lease areas to minimize ‘wind theft’ resulting from wakes generated by upstream wind turbine arrays. Uniquely detailed high-resolution simulations are performed with two different wind farm wake parameterizations to quantify power generation and wake-induced power losses from all offshore wind energy lease areas along the U.S. east coast. Annual Energy Production (AEP) from current leases is projected to be 139 to 173 TWh/yr. However, whole wind farm wakes extend over two to three times the footprint of existing lease areas. Those wakes from both local and remote wind turbines are projected to reduce the AEP by 15 to 49 TWh/yr. The simulation output is also used with new, robust, innovative tools to generate georeferenced data layers describing whole wind farm wakes (wind shadows) for use in planning and development. It is shown that wind shadows (wake reduction of wind speed) from the existing lease areas degrade the wind resource in up to 25% of the unleased area available after selecting areas with a good wind resource (mean wind speed >8.2 ms−1) and applying restrictions for water depth, distance to shore and to avoid busy shipping lanes. These results demonstrate the value of efforts to reduce wake losses to maximize power production efficiency. They also emphasize the importance of considering wakes in multi-criteria analyses to identify new lease areas for auction and for determining the resulting purchase price.

Caption: Wind climate at ∼ 150 m from observations and modeling. (a) 10 min observations from the LiDAR on the NYSERDA E05 buoy, (b) hourly output from the NOW-23 data set and (c) 10 min output from the WRF simulations presented herein at that location (40.1614°N, −72.7396°E). Wind speeds are discretized into classes that reflect different parts of the wind turbine power and thrust curves; 0–4 ms−1 where the blades are not rotating, 4–7 ms−1 where the thrust is high, 7–10 ms−1 with moderate thrust, 10–25 ms−1 where the thrust is relatively low and > 25 ms−1 when the wind turbines would cease rotation and power production. Probability distribution of (d) wind speeds, (e) wind directions and (f) turbulent kinetic energy (TKE) from the WRF simulations presented herein and the 21 year NOW-23 climatology from WRF at the centroid of the RIMA lease area cluster

Leave a Reply