Assessing atmospheric stability and its impact on rotor-disk wind characteristics at an onshore wind farm

As the average hub height of new wind turbine installations continues to increase, wind turbines typically encounter higher wind speeds, which enable them to extract large amounts of energy. However, they also face challenges due to the complex nature of wind flow and turbulence in the planetary boundary layer (PBL). Wind speed, direction and turbulence vary across a turbine's rotor disk in part on whether the PBL is stable, neutral or convective. A stable atmosphere generally has high wind shear and low turbulence while convective conditions are characterized by large amounts of turbulence and low wind shear. To assess the influence of atmospheric stability on wind characteristics in the rotor-disk, we utilize a unique dataset from a West Coast wind farm, including meteorological tower observations, surface eddy flux observations, and high-resolution measurements of wind speed and turbulence from a remote-sensing Doppler Sound Detection and Ranging (SODAR) instrument. We compare approaches to defining atmospheric stability, using direct measurement of the Obukhov length (L) as the standard or “truth”. Typical wind farm observations enable the calculation only of a wind shear exponent (α) or turbulence intensity (I_U) from cup anemometers at a few heights, while the presence of SODAR at this farm provides calculations of α and I_U at all heights within the rotor-disk and enables the direct observation of turbulence kinetic energy (TKE) profiles.

For each 10-minute period over the study year, we describe PBL stability conditions based on magnitudes of the Obukhov length (near-surface measurement), wind shear exponent (taken at various heights in the rotor disk), turbulence intensity (taken at hub-height), and turbulence kinetic energy (taken at hub-height). We classify each time period as belonging to one of five stability classes: strongly stable, stable, neutral (includes slightly stable and slightly convective), convective, or strongly convective. The SODAR-based parameters α, I_U, and TKE agreed well with the more physically-robust L, with TKE exhibiting the best agreement, while cup anemometer-based I_U could not accurately predict stability and greatly under-predicted convective conditions. Our results suggest that SODAR measurements not only quantify atmospheric stability with high accuracy, but also provide insight into wind speed and turbulence profiles in the rotor disk, which are likely to affect wind turbine power performance. Furthermore, standard wind farm meteorological measurements (e.g. cup anemometers) are insufficient for characterizing the performance of wind turbines in locations in which atmospheric stability is expected to have a significant impact on power performance. We suggest that observed wind farm “underperformance” may actually be due to inappropriate expectations of power based on inaccurate measurements from cup anemometers.