Measurements of inflow turbulence for large wind turbines

Wind turbine rotor diameters and hub heights have increased significantly in the last few decades and are projected to continue to increase, in an attempt to further reduce the cost of wind energy [9]. Consequently, wind turbines are beginning to operate in a part of the atmosphere about which our knowledge is limited [12]. Turbulence models such as the Mann [8] and Kaimal model [5] recommended by the International Electrotechnical Commission [1] for the design and certification of wind turbines may no longer be valid at these heights given that they were validated against measurements from relatively short met-masts. This may have a significant impact on the loads and cost-efficiency of modern wind turbines [12].
A number of studies have attempted to fill this knowledge gap. Cheynet et al. [4] compared measurements of spectra and vertical coherence from the FINO1 met-mast to the Kaimal model with Davenport’s exponential coherence. They observed the presence of mesoscale coherent structures even at heights of 81m above sea level. The COTUR experiment [3] used a novel approach to measure lateral coherence using pulsed lidars but the results were undermined by large uncertainty in the pointing directions of the beams. Forwarding-looking nacelle lidars were used at the HyWind Scotland wind farm [2] to measure lateral coherence between the line-of-sight component of the wind. The line-of-sight spectra showed high amplitudes at low frequencies further suggesting the presence of mesoscale structures within the hub heights of modern wind turbines. Inspired by these findings as well as the work of Lindborg et al. [7] and Larsén et al. [6], Syed and Mann [11] extended the Mann model to account for low-frequency mesoscale turbulence.
Thus, we present novel measurements of spectra and coherence of the true along-wind and cross-wind components at heights 150 and 250m above the sea as well as for lateral separations up to 200m which are then compared to the Syed-Mann model.