ECN publicatie: facebook
Stall Flexteeter: computations
Gepubliceerd door: Publicatie datum:
ECN 1995
ECN publicatienummer: Publicatie type:
ECN-C--95-010 ECN rapport
Aantal pagina's: Volledige tekst:
116  Niet beschikbaar.

In order to establish the applicability of aero-elastic codes PHATAS-2(ECN) and FLEXLAST (SPE)) for the design of flexible rotors, the results of a set of load and motion measurements were compared to computed loads and motions. The evaluation took into consideration 6 signals: the flap moment on the flexbeam, the lead/lag moment on the flexbeam, the torsional motion on the flexbeam, the flap angle, the teeter angle and the rotor shaft torque. Computations for rotor operation at a small yaw misalignment included 24 cases representing 8 different wind intervals, whereas the computations at a large yaw misalignment included 4 cases representing 4 different yaw intervals. This involved an evaluation of the aerodynamic modelling, the pitch mechanism, the blade mass and stiffness distribution, the teeter hinge, the generator and the drive train, and the tower. The rotor model was verified by comparing computed and measured rotor characteristics : the aerodynamic performance, the mechanical characteristics, and the teeter motion. It appeared that although the teeter motion is adequately modelled, there is no exact correspondence between the computed and the measured characteristics. The results of the two computations are mutually consistent. The results indicate up to the experimental uncertainty the average computed quantities agree with the corresponding measured quantities. The variation in the computed quantities, on the other hand, is smaller than the measured variation. For the yawed cases the same picture appeared although here large differences between the computed results were found. An evaluation of the power spectra of the computed and the measured quantities revealed these spectra have corresponding natural frequencies but different spectral densities. In particular, the computed magnitude of the second symmetric flap frequency is smaller than the measured magnitude of this frequency. This suggests the difference between the computed and the measured variations originates from large symmetric flap motions during turbine operation, which motions were not included in the input to the design codes. 40 figs., 21 tabs., 4 appendices, 24 refs.

Terug naar overzicht.