Turbulence Analysis Upstream of a Wind Turbine: a LES Approach to Improve Wind LIDAR Technology

Traditionally wind turbines learn about the incoming wind conditions by means of a wind vane and a cup anemometer. This approach presents two major limitations: 1) because the measurements are done at the nacelle, behind the rotor blades, the wind observations are perturbed inducing potential missalignement and power losses; 2) no direct information of the incoming turbulence is extracted, limiting the capacity to timely adjust the wind turbine against strong turbulent intensity events. Recent studies have explored the possibility of using wind LIDAR (Light Detection and Ranging) to overcome these limitations (Angelou et al. 2010 and Mikelsen et al., 2013). By installing a wind LIDAR at the nacelle of a wind turbine one can learn about the incoming wind and turbulent conditions ahead of time to timely readjust the turbine settings. Yet several questions remain to be answered such as how far upstream one should measure and what is the appropriate averaging time to extract valuable information. In light of recent results showing the relevance of atmospheric stratification in wind energy applications, it is expected that different averaging times and upstream scanning distances are advised for wind LIDAR measurements. A Large Eddy Simulation (LES) study exploring the use of wind LIDAR technology within a wind farm has been developed. The wind farm consists of an infinite array of horizontal axis wind turbines modeled using the actuator disk with rotation. The model also allows the turbines to dynamically adjust their yaw with the incoming wind vector. The flow is forced with a constant geostrophic wind and a time varying surface temperature reproducing a realistic diurnal cycle. Results will be presented showing the relevance of the averaging time for the different flow characteristics as well as the effect of different upstream scanning distances. While it is observed that within a large wind farm there are no-significant gains in power output by scanning further upstream, much can be learned about the incoming turbulence, hence allowing improved wind turbine readjustments. Time correlations with the upstream incoming turbulence have been computed through an entire diurnal cycle, and a non-dimensional analysis shows the existence of different behaviors throughout the day.