Servo control of the wind turbine under sinusoidal variation wind condition

[winder365]

Dear all,
I want to test the effect of servo control of the IEA-15 MW wind turbine operated a prescribed sinusoidal variation wind (same as the inflow condition of prescribed platform surge motion).

The wind is generated as V0 = 10.59+Asin(2πf*t), where A = 1.8 m/s, f=0.15 Hz. However the control effect is not very well.
As can be seen from the figure, power of the wind turbine may not be controlled very well and the power fluctuate significantly.

Thank you for your attention.
Best wishes,

[jjonkman]

Dear @winder365,

Can you clarify what you are asking?

Best regards,

[winder365]

Dear @jjonkman ,

Sorry for not clarifying the question.

I want to test the effect of servo control of the IEA-15 MW wind turbine operated a prescribed sinusoidal variation wind (same as the inflow condition of prescribed platform surge motion). A uniform wind file is generated as V0 = 10.59+Asin(2πf*t), where A = 1.8 m/s, f=0.15 Hz.

As can be seen in the figure, the power fluctuate significantly under such wind condition even activate the servo control (variable speed and variable pitch control are all activated through the Rosco controller for IEA-15 MW wind turbine).

So, I am confused that why the servo control of the wind turbine doesn't alleviate the fluctuation of the rotor power very well?

[jjonkman]

Dear @winder365,

By "with ServoControl", presumably you mean that you have GenDOF enabled in ElastoDyn and the baseline ROSCO controller for the IEA Wind 15-MW RWT enabled in ServoDyn, controlling pitch and torque; is that correct?

What have you set "without ServoControl"?

What "power" are you plotting? Is this GenPwr from ServoDyn or something else?

What are the rotor speed and blade-pitch angles doing for each case?

Best regards,

[winder365]

Dear @jjonkman,

Thank very much for you efficient response!

1. You are right! The GenDOF is enabled in ElastoDyn, and the baseline ROSCO controller for the IEA Wind 15-MW RWT enabled in ServoDyn. The relevant files are obtained for the github repository of the monopile IEA-15 MW. Both the pitch and speed of the rotor are controlled.

2. For without ServoControl, the GenDOF is set to false. And the wind turbine operated under with the rotor speed of 7.55 rpm, the wind is generated as V0 = 10.59+Asin(2πf*t), where A = 1.8 m/s, f=0.15 Hz.

3. The power I plot is the Aerodynamic power of the rotor (namely RtAeroPwr ), as I am more concerned to the performance of the rotor.

4. For without ServoControl condition, the rotor speed is constant 7.55 rpm, and the blade pitch is set to 0 deg. For with ServoControl condition, the initial speed is 7.55rpm, and the initial pitch is 0 deg, however this will be changed by the Rosco controller.

What I confused is that why the Rosco controller has such a little effect when the inflow wind fluctuates. The aerodynamic power of the rotor still fluctuate significantly after the Rosco controller is activated.

[dzalkind]

Hi,

Can you please also share the blade pitch, rotor speed, and generator torque of your simulation? Also note that PCMode in ServoDyn must be set when using the DLL-based controller.

Best, Dan

[winder365]

Dear @dzalkind @jjonkman ，

Thank you for your attention!

Of course, I have set PCMode to 5 (user-defined from Bladed-style DLL) in ServoDyn must be set when using the DLL-based controller.

The plot of blade pitch, rotor speed, and generator torque （I extracted the RotTorq） are shown as follows:

Best wishes,

[dzalkind]

Hi @winder365,

This looks like the natural response of the turbine to a sinusoidal wind input. You are essentially evaluating the closed-loop frequency response of the system with this simulation.

The frequency you are analyzing appears to be a difficult one to maintain the rated generator speed. At different (lower) frequencies, you might see better tracking. Compared to normal turbulence, the gust amplitude you are using is also very large.

I hope this helps.

Best, Dan