To provide full-scale experimental data on the performance of five wind turbine arrays located in a variety of configurations and environments and to compare with the predictions of existing numerical models.
Wind turbines sited in wind farms will normally experience reduced wind speeds and increased tubulence intensities compared with those operating in a freestream environment. This will results in the turbines generating less energy and also suffering greater fluctuating structural loads. The latter may result in accelerated fatigue damage rates.
This project is intended to provide full scale data on the wind fields encountered within wind turbine arrays and on the resulting array performance. Measurements are being made at 4 sites located in differing terrain. In addition, structural loads are being measured on selected turbines to enable the increase in turbine dynamic loads to be investigated and quantified.
Data are now available from all the sites in the project. A substantial body of data is now available, in a standard format, on the performance of wind turbine arrays in a number of terrain types. Although only preliminary analysis has so far been undertaken in most cases, a number of important features are apparent. The turbine power losses encountered deep within an array are more complex than suggested by current simple models. The measured turbine wake losses tend to saturate 2 or 3 machines into the array and, for smaller separations, may actually decrease as one passes further down a line of turbines. It appears that a reasonable prediction of wake losses in complex terrain may be obtained from a simple linear superposition of the wind speed perturbations arising from terrain and wake effects. The fluctuating loads experienced by a turbine, as measured by a damage equivalent constant stress, are significantly increased in a wind farm.
The validation of existing numerical models of the structure of wind turbine wakes and the performance of turbine arrays relies almost exclusively on data obtained from small scale wind-tunnel tests. Few comprehensive full-scale data are available on the characteristics of the flow within wind farms or on their overall efficiency. In addition, little information is available on the increased dynamic loads experienced by turbines when sited in an array.
This project is intended to provide full-scale experimental data from five wind farms located in various configurations and environments. The data will be used to validate existing numerical models and will also be utilized by two parallel JOULE projects ("Wake and wind farm modelling" and "Dynamic loads in wind farms"). The information is also expected to be of direct interest to machine designers and manufacturers.
Measurements will be made at the following sites:
. Sexberium (Netherlands) - uniform, flat site.
. Norrekaer Enge (Denmark)- flat, coastal site.
. Zeebrugge (Belgium) - linear array on harbour breakwater.
. Capel Cynon (U.K.) - moderately complex, upland site.
. Monte Ahumada (Spain) - complex, mountainous terrain.
Data on the mean velocity deficits and the intensity and spectral distribution of turbulence will be collected at selected locations within the arrays. Detailed measurements of the performance of individual turbines and the overall efficiency of the array as a function of both wind speed and direction will also be carried out. At some of the sites the dynamic loads experienced by a small number of selected turbines within the array will also be studied.
Funding SchemeCSC - Cost-sharing contracts