|
Title:
|
|
Variable speed turbine dynamic model validation. JWT measurements and simulations.
|
|
|
|
Author(s):
|
|
|
|
|
|
Published by:
|
Publication date:
|
|
ECN
Wind Energy
|
22-1-2007
|
|
|
|
ECN report number:
|
Document type:
|
|
ECN-E--07-008
|
ECN publication
|
|
|
|
Number of pages:
|
Full text:
|
|
70
|
Download PDF
|
Abstract:
In this report the Variable Speed Pitch (VSP) wind turbine dynamic model, developed by ECN and TUD in the Erao-2 project, is validated. Simulation results are compared to measurements of the JWT variable speed wind turbine in Sweden. Two types of measurements have been used: measurements during normal operation and a voltage dip. For the first set of measurements,
the Auto Power Spectral Density functions (APSDs) of measurement and simulation are compared. For the voltage dip, the time series of power and reactive power are compared.
The wind turbine dynamic model calculates instantaneous voltages and currents and includes the dynamics of the doubly fed indiction generator and its control but does not include the switching process of the power electronic converter. This is an argument for elimination of the high frequency noise on the measurements, for instance by calculating RMS or moving average values. Results with and without averaging are compared.
A complicating factor in this validation of the VSP model was the limited amount of information that was available for the JWT turbine, basically only the parameters in a commercial brochure. This is only a fraction of the parameters required by the dynamic turbine model.
To bypass this severe problem, the parameters of a similar VSP-DFIG were used in the validation. For this alternative turbine a complete and consistent set of parameters was available. Not knowing the parameters of the measured turbine, severely limits the validation results. A better set of variable speed turbine measurements was not available however, neither in the IEA Annex XXI data base nor at ECN-TUD.
The results for normal operation and voltage dip deviated on a number of points from the measurements, as could be expected. The deviations were sometimes relatively small and never extremely large. Within the limitation caused by the unavailability of turbine parameters, the validation results seem to be acceptable.
This report is the third of a set of three reports that documents the results of the project "Verificatie dynamische modellen van windparken (Erao-3)" and the K-Project "Validatie van windparkmodellen voor netstudies". The other two reports are entitled:
• Validation of dynamic models of wind farms (Erao-3): Executive summay, benchmark results and model improvements [6];
• Constant Speed Wind Farm Dynamic Model Validation: Alsvik measurements and simulations [5].
Back to List