When wind strikes an object—a building, a tree, a wind turbine—invisible atmospheric ‘wakes’ form downwind, similar to the water backwash behind a moving boat. In the case of wind farms, those wakes produce unwanted effects: reduction in wind speeds for downwind turbines and enhanced turbulence. Downstream turbines are less efficient than those first in line to catch the wind and experience greater damage from rapidly changing conditions and wake turbulence. These combined effects contribute to 10–20% underproduction of wind-turbine arrays as compared with the output of individual, isolated turbines.1 Wind-farm developers and operators need to understand important properties of these wakes (such as downwind extent, velocity deficits, and turbulence strength) in the real atmosphere to effectively design wind-farm layout and the turbines themselves.
Azimuth (conical-sector) scan at 3°
elevation, from Pichugina et al. (2011).2
This scan required approximately 17 seconds to complete, demonstrating how the High-Resolution Doppler Lidar (HRDL) can capture the rapid evolution of wakes. Vertical axis is distance north of the lidar, which is located at (0, 0), and horizontal axis is distance east. Doppler velocities are color coded (in m/s) as in the color bar at the top of the plot. The turbine is at the center of the white box. Black rectangles indicate locations where signal was intercepted by towers or turbines.
So far, designers have relied on untested assumptions and models to estimate how far wind wakes extend downstream and, therefore, how to space turbines on a farm. Information used to characterize wakes has been based on wind tunnel results, conceptual models, and numerical simulations, which give idealized conditions not representative of the atmosphere. Interactions among turbines and their wakes take place well above the earth's surface, at atmospheric levels, and are difficult to measure. Limited atmospheric measurements, generally from instrumented towers below the wakes, miss the complex wind dynamics aloft in the turbine rotor layer.
Doppler lidar is a laser remote-sensing instrument that overcomes the difficulties of accurately measuring winds above the surface. We developed a scanning Doppler lidar system that was recently set up at the National Renewable Energy Laboratory's (NREL) National Wind Technology Center (NWTC) test site near Boulder, Colorado, just upwind of a large 2.3MW wind turbine, which reaches 133m up into the atmosphere. This instrument, the High-Resolution Doppler Lidar (HRDL), was key in the Turbine Wake and Inflow Characterization Study (TWICS), a joint field program involving the National Oceanic and Atmospheric Administration (NOAA), the University of Colorado, the Cooperative Institute for Research in the Environmental Sciences, Lawrence Livermore National Laboratory, and NREL.
HRDL is a lidar that transmits a 200ns pulse of coherent 2.02μm IR radiation from a laser source employing thulium (Tm) and lutetium (Lu) in yttrium-aluminum-garnet at a pulse repetition frequency of 200Hz.3 The narrow collimated beam, less than 20cm in diameter at 4km, is sampled at 30m-range intervals. Averaging over 100 pulses yields ∼10–20cm/s precision in the radial, line-of-sight velocity measurement. Critical to its success is its full hemispheric scanning capability in azimuth and elevation. Azimuth sector scans through the turbine wake show its horizontal structure (see Figure 1). Elevation, or vertical-slice, scans illustrate the vertical structure.
The purpose of TWICS is to study turbine wake and inflow structure and properties under a range of environmental conditions, using data from HRDL and other instrument systems, and to evaluate computer simulations against the data. From the HRDL site at the NWTC, the scans mapped the wind field ahead of (inflow) and behind the turbine simultaneously: see Figure 1. The figure is an example of a nearly horizontal sector scan through the turbine wake (green), showing a clear wake deficit of 5m/s extending 600–700m downwind. On other occasions, we could identify wakes for distances of 2km or more from the turbine.
In all, we collected more than 130 hours of data, day and night, over 18 days during April and May 2011, over a wide range of wind speeds, stabilities, and inflow turbulence. The character of the wakes differed from day to night and from moderate to strong wind speeds. The data-analysis challenge for the data set will be to quantify the dependence of wake properties on environmental conditions to make the information available and useful to wind-farm designers.
We demonstrated the ability of NOAA's High-Resolution Doppler Lidar to provide needed measurements of the structure and characteristics of wakes behind a wind turbine. Understanding the effects of the turbulent lower atmosphere on turbines and turbine arrays is a critical obstacle to attaining the US Department of Energy goal of 20% renewable energy by 2030. Future work involves studying turbine wake effects under a wider range of ambient conditions and in other locations, and then extending to interactions of wakes from turbines in arrays. Under discussion is mounting HRDL and its motion-compensation system (used in deployments at sea) on a truck. This unit can then be driven back and forth within an operating wind farm to obtain detailed data on wake interactions.
We received financial support from the Renewable and Sustainable Energy Institute at the University of Colorado at Boulder and NREL, and from the US Department of Energy's Wind and Hydropower Technology Program. We appreciate the efforts of NREL staff in safely deploying the HRDL to NREL. We thank John Brown and Katy Human of the NOAA Earth System Research Laboratory for expert forecasting and editorial guidance, respectively, and Andy Clifton of NREL for data support.
Robert Banta, W. Alan Brewer, R. Michael Hardesty
National Oceanic and Atmospheric Administration (NOAA)
Earth System Research Laboratory
Cooperative Institute for Research in the Environmental Sciences
University of Colorado
National Renewable Energy Laboratory
National Wind Technology Center
Lawrence Livermore National Laboratory
1. R. J. Barthelmie, O. Rathmann, S. T. Frandsen, K. S. Hansen, E. Politis, J. Prospathopoulos, K. Rados, Modelling and measurements of wakes in large wind farms, J. Phys. Conf. Ser
. 75, pp. 012049, 2007. doi:10.1088/1742-6596/75/1/012049
2. Y. L. Pichugina, R. M. Banta, W. A. Brewer, J. K. Lundquist, N. D. Kelley, R. M. Hardesty, R. J. Alvarez, S. P. Sandberg, A. M. Weickmann, Wind turbine wake study by the NOAA High-Resolution Doppler Lidar, 2011. To be published in Proc. 16th Coherent Laser Radar Conf., Long Beach, CA, 19-24 June 2011.
3. C. J. Grund, R. M. Banta, J. L. George, J. N. Howell, M. J. Post, R. A. Richter, A. M. Weickmann, High-resolution Doppler lidar for boundary-layer and cloud research, J. Atmos. Ocean. Technol. 18, pp. 376-393, 2001.