OpenFAST
ROSCO Controller Question
Hello, I'm using the ROSCO controller the IEA 3.4 RWT, within this scenario:
Steady wind speed: 15.5 m/s
I set the initial conditions to: initial rotor speed: 12.1 rpm and initial blade pitch angles: 15.6 degress. Both are the expected values for a mean wind speed of 15.5 m/s at the Hub Height. I selected that values guided by this question: https://forums.nrel.gov/t/mach-number-exceeds-0-3/3375
I'm getting this results in the output power:
Is this correct? I don't know if it is not converging or this is the expected behavior. I tried to reduce the simulation timestep to 0.001, it really helps. Because using 0.05 the output power does not converge.
If I simulate using initial conditions in region 2: rotor speed: 5.0 pitch angle: 0.017 deg
the result looks like is converging better, but this error happens:
the time step is 0.001 as well.
What should I do to get convergency without errors?
Dear @vicemora,
Ignoring the warnings for now, I suspect you are not using the correct initial conditions of rotor speed and blade pitch. The forum post you linked to is not referring to the IEA Wind 3.4-MW reference wind turbine. According to Figure 17 from the IEA Wind 3.4-MW reference wind turbine report (https://www.nrel.gov/docs/fy19osti/73492.pdf), at a wind speed of 15.5 m/s, the rotor speed and pitch should be closer to 11.75 rpm and 14.5 degrees. Does that solve your issue?
Best regards,
@jjonkman
I changed the conditions to 11.75 rpm and 14.5 deegres. I'm using a timestep 0.001
The generator power is near the rated value (3.37 MW), but it still presents notable variations. Is that the expected behavior?
I'm trying to implement a controller where I need to know at each time step if the power is approaching or moving away from the rated value. So that, I think if the variant behavior of the power is normal, at each time step I might take another approach to determine wheter or not the power is approaching or moving away from the rated value.
@jjonkman For better understanding:
data3 is the pitch angle command and data4 the actual WT pitch angle, both in rad. You can see when I send another command the transient behavior is unexpected. It is not like a smooth aproximation but a immediate aproximation.
Dear @vicemora,
Regarding the oscillations in generator power, what is the frequency of oscillation? Is it related to one of the turbine natural frequencies? Is it tied to oscillations in rotor speed or generator torque or both?
Regarding the pitch angle, is this tied to the related questions you've been asking: https://github.com/OpenFAST/openfast/issues/1953?
Best regards,
Dear @jjonkman
I'm now very clear about the pitch angle issue.
About the oscillations, they happen only with speed, because I have set a cosntant torque in region 3. Is there a way to get rid of the oscillations. The oscillations are ruining my results, because in each step I have to know if the power is aproximating or getting away from the rated value.
I do not really know if it is related to the natural frequencies, I'm not very familiar with that topic.
Dear @vicemora,
What is the oscillation frequency of the generator speed? You can compare that to the natural frequencies of the IEA Wind 3.4-MW reference wind turbine, which are summarized in the following report: https://www.nrel.gov/docs/fy19osti/73492.pdf.
Best regards,