Effects of Atmospheric Thermal Stability on Wind Turbine Wake recovery: Volumetric scans with three simultaneous Wind LiDARs

The presence of wind turbine wakes can modify noticeably the natural transport phenomena within the atmospheric boundary layer (ABL), which could impact on the local climate and, in turn, on the physical and biogeochemical characteristics of the soil. Thus, the environmental impact of wind turbine wakes, especially in presence of large wind farms, might represent a hazard in case of surrounding terrains used for agriculture or farming. On the other hand, local atmospheric conditions, such as vertical profiles of the mean wind velocity and the turbulence intensity, which are in turn affected by the ABL stability regime, have a remarkable impact on the downstream evolution and recovery of wind turbine wakes. Therefore, mutual interactions between local atmospheric conditions and the wake flow induced by wind turbine rotation could take place. Moreover, evaluation of the effects of ABL stability on the wake evolution and, more interestingly, on wake-to-wake interactions within wind farms, is also of outmost relevance for optimization of wind farm layout and maximization of wind power harvesting. Driven by the above-mentioned considerations, a field campaign was performed in order to characterize the wind field produced by turbine wakes under different ABL stability regimes. Wind velocity measurements of the wake produced by a 2 MW Enercon E-70 wind turbine were carried out with three synchronized scanning Doppler wind Light Detection and Ranging (LiDAR) instruments. The main challenges in performing field measurements of wind turbine wakes are represented by the large measurement volumes involved, of the order of magnitude of few kilometers, and by the varying wind conditions and consequent adjustments of the turbine yaw angle, which are needed for maximization of power production. Different LiDAR scanning procedures were carried out in order to perform 2D and 3D characterizations of the mean wake velocity field. Classical Plan Position Indicator (PPI) scans were performed in order to characterize the downstream evolution of wind turbine wakes. These measurements are performed over conical surfaces intersecting the wake volume with different elevation angles. Those measurements present the advantage of allowing the characterization of the wake velocity field, even with a varying mean wind direction. However, it is difficult to perform a quantitative characterization of the wake recovery, because for different downstream locations the measurements are acquired at different heights, being the measurement plane inclined with a certain elevation angle. Another LiDAR measurement technique used to obtain 2D velocity fields is the Range-height Indicator (RHI) scan. For RHI scans the LiDAR elevation angle is swept over a certain range, while the azimuthal angle is fixed, in order to perform measurements of a vertical plane. This measurement technique is considered to be not very efficient, because of the rejection of a large amount of data due a possible misalignment between the measurement plane and the mean wind direction. Therefore, considering drawbacks connected with the PPI and RHI scans, a novel LiDAR scanning technique was performed by sweeping simultaneously elevation and azimuthal angles of the LiDAR, in order to scan over a 3D volume covering the whole wind turbine wake. With this technique, which is denoted as “volumetric scan”, a significant sampling time is typically obtained due to the large number of measured directions of the LiDAR laser beam. Consequently, the number of tested LiDAR directions is chosen as a tradeoff between the size of the measurement volume, the desirable spatial resolution, and the consequent sampling time. For our tests, measurements were carried out via two synchronized wind LiDARs located 12 rotor diameters, d, downstream of the turbine location. Each LiDAR measured over half of the measurement volume, producing a sampling period of 110 s for each volumetric scan. Simultaneously, another wind LiDAR was deployed at an upstream distance from the turbine location of 12d, in order to characterize the incoming wind. The mean velocity field connected with the wind turbine wake can be accurately characterized via volumetric scans, and specifically it can be evaluated over the horizontal plane at hub height and over the wake vertical symmetry plan, which are typically considered for the evaluation of the wake recovery. The minimum wake velocity deficit is estimated as a function of the downstream location, then fitted through a power law, whose exponent represents the wake recovery rate. It is found that stability condition of the ABL has a significant effect on the wake evolution; in particular the wake velocity deficit recovers with a higher rate under convective regimes. This result suggests that the environmental impact of wind turbine wakes can vary significantly for different ABL stability regimes, which should be also considered for improved wake models and predictions of wind power harvesting. Besides LiDAR scanning techniques designed for the characterization of the mean velocity field of wind turbine wakes, other tests were devoted to the direct characterization of wind turbulence. Those measurements are typically performed by staring the LiDAR laser beams over fixed directions for a certain sampling period, and by acquiring data with the maximum sampling frequency. Staring measurements performed with a single LiDAR already allow the characterization of temporal fluctuations of the measured radial velocity. Specifically, when this measurements are performed within the wind turbine wake, an area with a higher turbulence level is detected in proximity of the top tip of the turbine rotor, which confirms previous results obtained via wind tunnel tests and Large Eddy simulations. This enhanced turbulence level has a mechanic origin, and it is connected to significant wind shear due to the overlap of a typical ABL vertical velocity profile with a Gaussian-like velocity field produced by a wind turbine wake. The LiDAR staring measurements also allow an accurate characterization of the inertial sub-range of turbulence. A more sophisticated scanning technique, consisting in three simultaneous intersecting staring LiDARs, was performed in order to obtain 3D turbulence measurements. This scanning technique was assessed via sonic anemometer data. It is show that accurate 3D wind turbulence measurements are obtained with this measurement technique, and in particular they can provide novel and important features related to the dynamics of wind turbine wakes.