How to best simulate a fixed-speed active-stall-controlled wind turbine

Wind & Water Computer-Aided Engineering Software Tools
1

Dear all,

We are currently looking into retrofit options for older fixed-speed active-stall-controlled wind turbines using lidar-assisted-control. To evaluate the potential, we would like to perform a simulation study with the NREL 5MW reference wind turbine.

We are using the OpenFAST sFunction with Lidar Simulator based on https://github.com/sowentoDavidSchlipf/openfast/tree/f/lidarsim).
For the airfoils we use the “smoothed” ones kindly provided by Jason:
NREL 5MW Rotor Geometry - #14 by Dawn.Ward

For the power regulation, we would like to use the measured electrical power to adjust the pitch angle to negative pitch angles, when rated power is reached.

For this purpose, we disabled the Generator DOF (GenDOF) and did the following adjustments in the ServoDyn input file:

  • set VSContrl to 0 (none)
  • set GenMode to 1 (simple induction generator)
  • set the values for the simple induction generator based on Jason’s definition [url]http://www.doi.org/10.2172/947422[/url] to
---------------------- SIMPLE INDUCTION GENERATOR ------------------------------
10          SIG_SlPc    - Rated generator slip percentage (%) [used only when VSContrl=0 and GenModel=1]
1067        SIG_SySp    - Synchronous (zero-torque) generator speed (rpm) [used only when VSContrl=0 and GenModel=1]
43093.55    SIG_RtTq    - Rated torque (N-m) [used only when VSContrl=0 and GenModel=1]
2           SIG_PORt    - Pull-out ratio (Tpullout/Trated) (-) [used only when VSContrl=0 and GenModel=1]

Here, the signal GenPwr is then constant 5 MW, even when the wind is very low. We cannot use this then for active-stall control.

Currently, we are using following work-around:

  • Multiply the high-shaft-power HSShftPwr with the generator efficiency and use the resulting signal for control
  • We also had to disable the Drivetrain rotational-flexibility DOF (DrTrDOF) and blade modes, since with these modes enabled, the fluctuations have been extremely high.

Here our question: Is there a better way to simulate a fixed-speed active-stall-controlled wind turbine?
Any idea would be very much appreciated!

2

Dear David,

Just a few comments:

  • Disabling the generator DOF in ElastoDyn means that the generator will spin at a fixed speed regardless of the torques (aerodynamic, generator) applied. In this case, the torque controller is quite meaningless because it has no effect on the generator speed anyway. This is probably not want you want, unless that limitation is satisfactory for you.
  • Even a fixed speed wind turbine likely has a small amount of speed and torque variation, which, you can simulate, e.g., with the SIG model as you have done. The smaller the slip percentage, the steeper the torque-speed curve will be, and the steeper the torque-speed curve is, the more the speed will remain fixed. (FYI: A small integration time step may be required if the torque-speed curve is very steep because a very steep torque-speed curve can effect the numerical stability of OpenFAST.)
  • We implemented an active pitch-to-stall controller for the NREL 5-MW baseline wind turbine in the following paper: nrel.gov/docs/fy08osti/42589.pdf. In this paper, generator speed wasn’t fixed; instead, we used the same torque controller used in the active pitch-to-feather controller (which used constant power in Region 3, rather than constant speed in Region 3).

I hope that helps.

Best regards,

3

Thanks Jason, this helped!

It is working now with following settings:

  • Generator DOF and all the other modes (DrTrDOF and blades) enabled:
  • VSContrl = 0
  • GenMode = 1 and the values for the simple induction generator above. The slip percentage of 10% might be a bit high (from a smaller, real turbine it seems more like 1%) and also the synchronous generator speed does not fit to 50 or 60 Hz grid, but it is a made up problem anyway.
  • Using GenPwr for the active-stall controller

The rotor speed and electrical power are fluctuating quite a lot. I tested also 1%, but then I got huge oscillations for higher wind speeds.

Btw, is there any OpenFAST model available for older fixed-speed, active-stall-controlled wind turbines?

4

Hi David,

I’m glad that worked.

Sorry, but NREL has not released a model in the public domain of an older fixed-speed, active-stall-controlled wind turbine.

Best regards,

5

Dear Dr Jason
Please i want to simulate a wind 5mw wind turbine onshore and a problem is occurred
can you help me to fix this attached problem

Capture fast.PNG
Capture fast.PNG1860×523 24.5 KB

6

Dear Ali,

The first error appears to be caused by a formatting problem in your OpenFAST primary (*.fst) input file. As with any input file processing issue, I would enable the Echo option to debug. Hopefully resolving this error will solve the remaining errors.

Best regards,

7

Dear all,

just realized that somehow a “)” ended up in the link to the lidar simulator, sorry. Unfortunately, I could not change it above. So, the correct link is

Further, there will be a paper at the Torque 2022 about an update including some new features (blade blockage, unfrozen turbulence and flexible data availability).

Best regards
David