| Correct Tower Distortion Window | |
The Correct Tower Distortion window applies the method specified in Annex S of the IEC 61400-12 (2017) standards to correct for some of the effects on side-mounted anemometers of flow distortion around a lattice tower. To access this window choose from the menu.
The process begins by calculating a linear least squares regression between the valid concurrent speeds reported by the primary and control anemometers, to find the best-fit slope and offset in this relation:
where: | ||
 | is the regression-predicted wind speed for the primary anemometer | |
| U2 | is the wind speed from the control anemometer | |
| m | is the best-fit slope | |
| b | is the best-fit intercept |
Windographer then calculates the resulting residuals by subtracting the regression-predicted primary anemometer speeds from the actual speeds reported by the primary anemometer:
The window displays a scatter plot of these residual values versus direction, including values flagged with the special tower shading flag. Windographer then fits a sine curve to the residuals (excluding the tower-shaded values) to find the best fit phase and amplitude parameters in the following equation:
where: | ||
| A | is the amplitude parameter | |
| C | is the phase parameter (identified in the IEC standards as Centre) |
Windographer uses a grid search approach to find the phase and amplitude parameters that minimize the mean squared error of the sine curve compared to the whole set of residuals.
Tip: This sine curve fit process may use fewer time steps than the linear least squares regression described above, because the sine curve fit process requires valid data in the wind direction sensor as well as in the primary and control anemometers.
Having calculated the best-fit sine wave, Windographer then applies the following corrections to the primary and control anemometers (mean, min, and max columns) to remove the effect of tower distortion outside the shaded sectors:
where: | ||
| U1 | is the original primary anemometer speed | |
| U1,corr | is the corrected primary anemometer speed | |
| U2 | is the original control anemometer speed | |
| U2,corr | is the corrected control anemometer speed |
Finally, Windographer adjusts the primary and control anemometers' associated standard deviation columns by scaling each value by the ratio of corrected speed to original speed, meanings the tower distortion correction leaves turbulence intensity unchanged:
where: | ||
| USD | is the original speed standard deviation value | |
| USD,corr | is the corrected speed standard deviation value | |
| U | is the original speed value | |
| Ucorr | is the corrected speed value |
In the residuals graph you can choose to display the residuals after the sine wave correction, so you can see the imminent effect of this correction. You can choose to apply the correction to all pairs of colocated anemometers in the entire period of record, or just a subset.
See also