PDF《wind direction changes (EOD)》

2], where the value

1 indicates no correlation between the maximum wind speed gust and the wind direction gust, and the value

2 indicates full correlation between these. The maximum GDI value relates to the ECD gust duration. In the present investigation four different duration periods defined by Δt=2, 5, 10, and

30 seconds, respectively, has been determined for each 10-minute period using a moving window technique. Figure

1 shows an example of a gust event, recorded in complex terrain. The figure illustrates how a negative wind speed gust (16m/s → 8m/s → 16m/s) correlates with a wind direction gust (320 deg. →365 deg. →320 deg.) during a gust duration time span of

5 seconds. Values of Δt ≥

5 seconds typically represent a gust covering a 100m diameter rotor for high mean wind speeds. In the present analysis GDI values larger than 1.98 are used as representative for fully correlated wind speed- and wind direction gust situations. The GDI range limit equal to 1.98 has been selected to obtain a reasonable amount of these compound gust events. A GDI value equal to 2.00 turned out to result in very few recordings satisfying this requirement, . ))) ( ) ( ( max( )) ( ) ( ( ))) ( ) ( ( max( )) ( ) ( ( t Dir dt t Dir abs t Dir dt t Dir abs t U dt t U abs t U dt t U abs GDI ? + ? + + ? + ? + =

400 410

420 430

440 450

460 470

480 490

500 5

10 15

20 25 Site=Oak Creek, h=10m;

22/3-2000;

22:58-23:08 WS (h=10m) - m/s

400 410

420 430

440 450

460 470

480 490

500 300

320 340

360 380 wd (h=10m) - deg Time - seconds Coherent gust and wind direction change Figure 1: Example of a measured synchronous wind speed gust and wind direction change. 4. Measurements Database on Wind Characteristics [2] contains time series of measurements from more than

60 sites, representing a variety of different terrain types as documented in [4]. Based on huge amounts of

10 minute registrations, the probability of such joint gust observations, referring to different site types, has been shown on Figure 2. The figure indicates a high occurrence probability (PDF-value >

0.001) for a number of sites, but unfortunately not all sites are suitable for this analysis due to the vertical spacing criteria (Δh≤1 m) for wind speed- and wind direction sensors, respectively, and furthermore some sites includes partly disturbing wind turbines, ........