Detailed knowledge of the wind resource is critical for all aspects of a wind project, from exploration and financing to operation. Cup anemometers mounted on meteorological (met) towers are traditionally used, but as turbines grow, masts for these instruments have become taller and more expensive to build. Hence, remote sensing (RS) techniques such as LIDAR and SODAR are increasingly used for resource assessment. [28]
Reduced Risk
Detailed knowledge of the wind resource is essential for any wind farm development. However, the physical limitations of masts that mount cup anemometers have led to the development of remote sensing (RS) techniques such as LIDAR and SODAR, which use Doppler effect technologies – electromagnetic radiation in the case of LIDAR and sound in the case of SODAR – to measure wind speed and direction without the need for a mast.
This study compared the performance of these two commercial RS instruments to measurements from an instrumented mast in complex upland terrain, typical of many new and emerging sites for wind energy worldwide. When plotted as a function of height, both the RS wind speed and direction data showed considerable scatter with the cup anemometer-derived wind. Filtering based on internal instrument parameters that reflect the quality of the data collected during each 10-minute averaging period resulted in a much-reduced scatter and significantly improved correlations, particularly for the SODAR.
Increased Flexibility
LIDAR and triton sodar tier iii certified transmit energy pulse into the atmospheric boundary layer, reflecting off particles. These returns’ intensity and Doppler shift can then be used to determine wind speed and direction.
An intercomparison study between a US SODAR design that is now commercially available and an instrumented met mast was performed in upland terrain in the US [28]. Unfiltered RS data compared to cup anemometer wind speed and direction showed good agreement for up to 80 m. When filtered by obstacle and wind quality parameters, the correlations between the RS instruments and the mast data were much improved. This demonstrates the importance of filtering, even for RS systems that have been designed to be accurate without the use of mast-based data.
Increased Accuracy
Detailed knowledge of a site’s wind resource is necessary for most project phases, from site selection to development and operation. As physical limits on the size of masts for mounting cup anemometers limit their use, remote sensing (RS) technologies that can monitor turbulence structure and wind speed without needing a tall, instrumented met mast have emerged.
The two most common commercial RS instruments are LIDAR and SODAR, which utilize Doppler effect techniques to measure wind speed using light or sound. In a recent intercomparison study of LIDAR and SODAR data at a single 80-m height, the root mean square error between LIDAR and a standard cup anemometer was less than 2%.
However, the comparison plots in this study demonstrate that both RS instruments exhibited scatter in their data, particularly at lower wind speeds and with an easterly wind direction. These scatters largely reflect obstacle and mast effects, which can be filtered out.
Reduced Downtime
In upland terrain, where many wind energy sites are located, constructing traditional cup anemometer-based masts is often difficult and costly. However, remote sensing (RS) techniques can obtain wind measurement data without such masts.
LIDAR and SODAR use the Doppler effect to measure air movement in the atmospheric boundary layer, relying on either the reflected electromagnetic radiation from particles or the echoes from the varying temperature structure of the atmosphere. They can determine wind speed and direction and are particularly useful in measuring turbulence intensity.
For this study, eight Vaisala Triton Wind Profiler instruments were co-located with AQ500 SODARs and ZephIR wind meters to capture detailed measurements of horizontal velocity at 30 m, 50 m, 63 m, 80 m and 200 m above ground level every 10 minutes. The resulting data were subjected to internal quality filtering for both the LIDAR and SODAR (Points in Fit (PiF) and Packets in Average (PiA) for the LIDAR, Signal-to-Noise (SNR) for the SODAR). The unfiltered 80 m wind speed residuals shown in Figure 7(a,b) are comparable to those of the ZephIR, although there are periods when they under-record cup anemometer wind speeds.
