Hi,
I have been testing the Rosco controller with OpenFAST using the simple yaw rate control Y_ControlMode = 1 in the DISCON-ROSCO.IN file (attached zip file). I am also applying a wind profile with a varying wind direction from -45 to 45 deg matching (hopefully) the controller Y_Rate.
Basically, what I would expect to see is the actual yawing of the turbine with a rotor skew angle perhaps a bit constant across time. I am observing a profile more like this:
Also, looking at the vtp outputs I cannot see the actual yawing (over time) of the turbine just running the animation view. I was wondering if that something I am suppose to visualize given the actual yaw rate. I tried to plot the OrientationZ of the Hub over time (with left screenshot of the turbine and the end time - no rotated) but I am not clear is there is an actual yaw there:
Please any clarification will be appreciated.
Regards,
Juan
Yaw_Control_NWTC.zip (3.63 MB)
Dear Juan,
I haven’t reviewed your files in detail, but I did notice that while you’ve set YCMode = 5, you have not set the start time to enable yaw control until well after the simulation ends (TYCOn = 9999.9 s). Does setting TYCOn = 0.0 solve your problem?
Best regards,
Thanks Jason. That was it, I can see the yawing happening now.
Just an extra question, what would be the reason for the turbine to stay (after about 30 secs in this simulation) gyrated at 180 degrees?
Regards,
Juan
Dear Juan,
Sorry, but I’m not sure I understand what you are asking.
Best regards,
Sorry Jason, what I meant if there is another setting (probably in the yawing controller?) which stops the turbine at 180 degrees? I noticed running the files that the rotor stays inline with the wind for about the last 20 secs.
Regards,
Juan
Hi Juan,
I’m glad to hear you are working with the yaw rate controller. To answer your primary question: there is nothing in ROSCO to stop the yaw controller at 180 degrees.
One question: where did you get your yaw rate control parameters? This code was originally a part of the DRC-Fortran code, before it was adapted to become ROSCO. You can see the parameters they used here:
github.com/TUDelft-DataDrivenCo … CON.IN#L68
Additionally, I was able to replicate your results on my end. I haven’t dug into it too much, but it looks like something may be up with your input files and your inflow wind direction is actually 135 degrees off, not 45 degrees off.
It looks like the yaw controller is working as expected, but the initial transients you are seeing is simply because the turbine has to rotate so far.
Hope this helps,
Nikhar
Hi Nikhar,
Thank you for your answer.
In regards to the control parameters, I chose the yaw rate “Y_Rate” (i.e. 7.2 deg/s or 0.126 rad/s) to match somehow the wind direction rate, from 315 degrees to 135 degres. Can I confirm with you that this should follow the same clockwise orientation as the yaw rate? When you mentioned the wind direction is off, can you point me to how did you noticed it?
For the other parameters, I guessed some values as per the suggested defaults (so they were above zero):
0.01 ! Y_ErrThresh - Yaw error threshold. Turbine begins to yaw when it passes this. [rad^2 s]
0.107 ! Y_IPC_omegaLP - Low-pass filter corner frequency for the Yaw-by-IPC controller to filtering the yaw alignment error, [rad/s].
0.01 ! Y_IPC_zetaLP - Low-pass filter damping factor for the Yaw-by-IPC controller to filtering the yaw alignment error, [-].
1.0 ! Y_omegaLPFast - Corner frequency fast low pass filter, 1.0 [Hz]
0.017 ! Y_omegaLPSlow - Corner frequency slow low pass filter, 1/60 [Hz]
I will try to use the values from DRC to check if it helps to dissipate the transients.
Regards,
Juan
Hi Juan,
It is a bit surprising to me that both the x and y wind directions ended up negative. Particularly, I would have expected the x-direction to stay positive, and the y direction to shift positively or negatively (depending on +/- 45 degree change). Incidentally, the absolute yaw position shifts to -135 degrees, rather than +/- 45 degrees.
Hopefully the updated parameters help smooth the responses.
Cheers,
Nikhar
Dear Juan,
You mentioned that you applied a wind profile with a varying wind direction from -45 to 45 deg. Did you use the turbsim to produce the wind file which the wind direction is varying. If that, could you tell me how to set in the Turbsim?
Thank you in advance,
Jiaping.Cui
Dear @Jiaping.Cui,
You can not use TurbSim to generate wind data files with large variations in wind direction. My guess is that @Juan.Nunez made use of a uniform wind data file (WindType = 2 in InflowWind).
Best regards,
Dear Jiaping,
Yes, Jason is right. I used a uniform WindType with InflowWind with a .wnd file I created with the varying wind direction (i.e. Delta, positive clockwise looking down). Below couple of screenshots so you have an idea if it helps.
Regards,
Juan
Dear Jason and Juan,
Thanks for your explanation, and I understand that.
Best regards,
Jiaping.Cui
