Hi Dan,
Many thanks for your reply and insight. I do not quite understand why setting Fl_Mode to 1 gives issues (i.e. excessive pitching / instability, see 15MW towerbase pitching moment discrepancy - #9 by Salem.Okpokparoro ) for my model. Any take on this?
Regards
Salem
Hi Salem,
Fl_Mode of 1 uses the nacelle IMU translational acceleration (in the x-direction) to provide floating feedback. The gain Fl_Kp will translate this into a blade pitch signal.
Fl_Mode of 2 uses the rotational acceleration (about the y-axis) to determine the floating feedback pitch term. The same Fl_Kp gain is used to translate the rotational acceleration into a blade pitch signal.
So, if the same Fl_Kp is used in both modes, it will be too high when Fl_Mode is 1 by a factor of the tower height. Fl_Mode of 2 is the default and re-tuning using the ROSCO Toolbox should account for these factors if you choose to use Fl_Mode of 1.
Best, Dan
Hi Dan,
Thank you for your responses, much appreciated.
Regards
Salem
Dear all,
I’m trying to develop a new strategy for tuning the pitch controller of FOWT (PI regulator + parallel compensation). The preliminary results are promising: the controller shows improved rotor speed regulation under turbulent inflow conditions (U = 12 m/s), maintaining system stability while enhancing power tracking performance.
In the figure below, the blue curve represents the baseline (detuned) controller, while the red curve corresponds to the new controller I’ve developed (referred to as FF):
To further evaluate performance, I analyzed the power spectral density (PSD) of the rotor speed signals, comparing the detuned and FF controllers. For each controller, the PSD was computed both for the flexible floating turbine and a rigid configuration:
The results indicate that the FF controller significantly reduces rotor speed sensitivity to low-frequency turbulence, which is the main reason for its improved tracking performance. However, there’s a specific feature in the spectrum that caught my attention.
While most resonance peaks are clearly visible, the dynamic amplification around 0.07 Hz is less obvious. This peak appears much more pronounced in the flexible turbine configuration. Given that all known flexible modes of the turbine lie at higher frequencies, it’s unlikely that the 0.07 Hz peak corresponds to a structural mode.
I’m wondering whether this could be related to the platform pitch motion. Specifically, the platform pitch natural frequency is around 0.035 Hz, so it is possible that the 0.07 Hz feature is a harmonic (2×) of that motion.
Has anyone encountered a similar phenomenon? Could this be an interaction between the controller dynamics and second-order effects from platform pitch? Any insights or references would be greatly appreciated.
Thank you in advance!
Best regards,
Marco