Problems with ROSCO controller with 15 MW wind turbine

Wind & Water Controls
1

Hello,
I got a problem by using the Simulink model of ROSCO, which can be found on Github ( GitHub - NREL/ROSCO: A Reference Open Source Controller for Wind Turbines).
The problem is that the steady-state values that I get from the Simulink model (containing the S-function that calls OPENFAST) are not the same values that I can find in the paper of ROSCO (https://wes.copernicus.org/articles/7/53/2022/wes-7-53-2022.pdf).
The Simulink model is the original one without modifications.
the steady-state values are taken from the output of FAST, the wind is steady without shear effect and the precone and tilt angle is set to null. Only the rotor degree of freedom is enabled.
What could be the cause of the problem? Especially for the rotor speed.
thanks in advance for your support,
Best regards
image
image
image
image

image
image1087×421 120 KB

2

Hi Massimo,

I would take a closer look at the simulations near rated wind speed, at 10 and 11 m/s. Sometimes, they can be unstable in steady wind simulations as we tune the ROSCO parameters for turbulent conditions. If you are only concerned about the steady-state responses, you can reduce the PI pitch control gains and usually fix this.

I should note that the Simulink model isn’t supported as well as the FORTRAN version of ROSCO and is not an exact match. If you have any improvements to suggest, they would be greatly appreciated by the community.

Best, Dan