1 Fluid Mechanics, Department of Mechanical Engineering, Technical University of Denmark2 Department of Mechanical Engineering, Technical University of Denmark3 Department of Wind Energy, Technical University of Denmark
ELSAM: VU106, MEK-FM 2006-01
Modern wind turbines have reached a size that makes hub height wind speed measurements rather expensive. The cost of masts increases rapidly with height (distinctly more than linearly) and their installation is subject to a (often) lengthy authorization procedure. A ground-based SODAR is able to measure at many levels simultaneously and is economically competitive with other forms of measurements. The main objectives of the project have been to optimize the SODAR availability and the quality of the SODAR readings. Based on the SODAR readings the reliability, accuracy, limits and limitations of the SODAR have been determined. A validation program with a SODAR has been performed for 1½ year at a remote location without any access to an electricity grid. The program will study whether the SODAR wind measurements are accurate enough for wind power assessment and whether the SODAR is applicable as a stand-alone instrument or as a “profiler” in combination with reference instruments on a 30-50m tower. Remote operation of the SODAR system was tested against traditional wind resource measurement set-up equipment. The test set-up consisted of a 50 m mast, equipped with three calibrated cup-anemometers, 1 wind vane and one 3-D ultrasonic anemometer sampled at 10 Hz using a low- power industrial computer. Statistics were transmitted every 10 minutes using SMS telegrams. The SODAR was a METEK PCS-2000.24 low power version. All equipment was powered by a commercial off-grid power system designed to continuously provide 120 watt. The equipment was installed on gently rolling, open farmland with few windbreaks. A SODAR provides approximately 20 instantaneous scans every minute. This is much less than a traditional cup anemometer / data logger setup. In addition, the instantaneous values do not have any inherent temporal averaging (rather vertical spatial averaging) and are often noisy. For good statistical reliability it is necessary to compare at least ten-minute average values. Good agreement between cup and SODAR average values is first seen when comparing 1-hour average values. Comparison between the SODAR and reference cup-anemometer readings revealed slightly higher indications (less than 0.1 m/s) in favor of the cup-anemometers and a corresponding averaged power law exponent of 0.26. Since the SODAR is an acoustic instrument, the operation is limited by the background noise, which results in a lack of valid measurements for wind speeds above 14 m/s. - The findings have been reported on the European Wind Energy Association Conference in Athens 27 February – 2 March 2006 (Annex F). - Three potential master thesis projects have been identified based on the findings. - A complete dataset, consisting of SODAR and cup-anemometer measurements have been compiled, this is available on request (Annex D).