Dear Arash,
That all sounds correct.
Best regards,
Dear Jason,
Thank you for your constant support.
I am trying to find aerodynamic damping ratio from free decay simulation. I used to calculate aerodynamic damping from the initial period ( Usually after removing two first cycles ).
Now I tried to do another approach. What I am doing now is to apply uniform wind till time=40s and then for 4 sec I increase wind speed by 50%, then after time =44 sec again wind speed comes back to its constant value and then I calculate damping ratio because of this pulse wind speed.
The thing that is interesting is that for wind speed less than 12m/s, the results are similar to the ones I get from initial period (after removing first two cycles), but for wind speeds more than 12 m/s, in some wind speed, literally there is negligible damping and for some wind speed there is high damping.
To illustrate, I uploaded the time history plot for 16m/s, 18m/s and 20 m/s. You can see for 18 m/s, vibration dies out quickly, but for 20m/s and 16m/s, it almost does not dampen at all. Also, I uploaded the aerodynamic damping obtained from two approaches (Figure A).
Please find the figures here (app.box.com/s/rjdn2r9jc24damv2yr3wuqq2e6317zgh)
What do you think is the reason that the results differ this much? Please note that all situations are the same, I just changed wind speed (uniform wind speed).
Thank you very much for your help.
Best Regards,
Arash,
Dear Arash,
I agree that the results look odd, but I’m not sure what the problem might be. One guess is that the controller is not behaving as you’d expect–have you looked at the rotor speed, pitch angle, and generator torque to see if the responses are reasonable during the transient? It is hard for me to identify the problem without knowing anything about your model. Perhaps you should run the same study with the rotor speed spinning at a fixed rate to see if the results are as you expect.
Best regards,
Dear Jason,
By the aerodynamic damping, are you referring to the effects because of “structural velocity of blade” and “structural velocity of tower” which are considered in the calculation of “resultant velocity” seen by blade which in turn is used for aerodynamic force calculations? This may have influence on blade oscillations but not so much on tower oscillations. Right?
or , the tower aerodynamic loads (i.e since tower is oscillating in wind, it experiences drag force as a distributed load) ? Also, when we specify damping ratios (zeta) for fore-aft or side-side modes in FAST, how are the damping Coefficients (c) calculated inside FAST? using the formula c= 2zetamass*omega_n?
Thanks.
Regards,
Kumara
Dear Kumara,
In my response to Arash, I was referring to the aerodynamic damping of the rotor, which results from a combination of blade, tower, and wind velocities. For most wind turbines operating normally, the aerodynamic damping from the rotor dominates over the aerodynamic damping from the tower, at least in the fore-aft direction (along the wind direction). But the aerodynamic damping of the rotor has a strong impact on the tower response.
Yes, your equation for the tower structural damping (c) is correct, except that FAST uses the equivalent expression in terms of stiffness, rather than mass:
c = 2zetastiffness/omega
Regarding this last point, see my post dated Oct 08, 2013 in the following topic for more information: Natural frequency and damping ratio calculation - #7 by Jason.Jonkman.
Best regards,
Dear Jason,
The link you mentioned is very helpful. Thanks alot.
Regards,
Kumara