Hi Moderators,
I am trying to implement the controls developed in “Advanced Control Design for Wind Turbines Part I: Control Design, Implementation, and Initial Tests” by Wright, A D on test18 (FAST Certification Test #18: NREL 5.0 MW Baseline Wind Turbine (Onshore)). I have few questions.
1)How to use the input wind files available in the appendix A of the report and what changes should I make in the InflowWind v3.01 which is by default WindType=3
2)Is there any option available to print two parameters against each other (let’s say attached generator torque vs generator speed) or I have to do it manually in Matlab?
Regards,
Kamran Ali
Dear Kamran,
The wind data files shown in Appendix A of Wright’s report that you reference is in the format of InflowWind Uniform Wind File (WindType = 2).
The image you shown cannot be directly output from FAST, but can output both generator torque and generator speed so that you make such a plot yourself.
Best regards,
Hi Jason Jonkman,
Now, I am trying to realize “Region 3 Collective Pitch Control Design Example” in the above-mentioned manual (“Advanced Control Design for Wind Turbines Part I: Control Design, Implementation, and Initial Tests”) Wright, Fingersh - 2008 - Advanced Control Design for Wind Turbines Part I Control Design , Implementation , and Initial Tests.pdf (1.99 MB) on test 18 and for that, I am trying to linearize it first but I am confused about finding the operating point for linearization.
In the manual (page45) RotSpeed is set to 41.7 rpm, the BlPitch(1) and BlPitch(2) values to 11 degrees to reflect the initial conditions for the linearization simulation, how should I compute these points for test 18 or any other test, considering the fast linearization (times at which to linearize)?
Dear Kamran,
One would normally want to linearize about a steady-state condition. For the NREL 5-MW wind turbine, steady-state values of rotor speed and blade-pitch angle as a function of steady wind speed are published in the NREL 5-MW turbine specifications report: nrel.gov/docs/fy09osti/38060.pdf. So, set the steady wind speed you want linearize about, disable the blade-pitch controller and fix the blade-pitch angles at the appropriate angle for that wind speed, initialize the rotor speed to the appropriate speed for that wind speed, run the time-domain simulation until all start-up transients have been eliminated (the rotor speed should basically remain unchanged), and linearize the model at that time.
Best regards,
Thank you for your prompt reply.
Yes, I understand that I have to choose the operating point in the steady-state conditions. I have partly got an answer to the issue from the report (Definition of a 5-MW Reference Wind Turbine for Offshore System Development).
How do I have to compute the linearization time? (variable LinTimes in the input file)
Because the plots in the report mentioned are against the wind speed, and once done how do validate my linearised system.
Thank you so much once again for your assistance.
Best Regards,
update on my last question
I run the fast under the condition you have mentioned (i.e. set the steady wind speed you want to linearize about, disable the blade-pitch controller and fix the blade-pitch angles at the appropriate angle for that wind speed, initialize the rotor speed to the appropriate speed for that wind speed) and plotted the output using PlotFASToutput . I could see states achieving steady state after a certain amount of time. So from there, I can find out the time to linearize.
I choose rotor speed as 11rpm, steady wind speed as 11 (HWindSpeed) and blade pitch at 11 degrees as a test, but I am facing opposite to what you have mentioned about the rotor speed.
![]()
![]()
Dear Kamran,
Your process is fine, but from the steady-state response curves of the NREL 5-MW turbine in the report I linked to, you’ll see that at a wind-speed of 11 m/s, the blade-pitch angle should be 0 degrees and the rotor speed should be close to 12 rpm. Your results show that at 11 m/s and a blade-pitch of 11 degrees, the steady-state rotor speed is around 13 rpm, but this is not a normal operational condition of the NREL 5-MW turbine.
Best regards,
Dear Jonkman,
Thank you for your reply.
I am trying to linearize at blade pitch angles at 10, rotor speed 13 and wind speed at 15 in relation to the steady states values of reference report. Rotor speed is still changing. Where do you think, I am going wrong?
Dear Kamran,
At a steady wind speed of 15 m/s, the blade-pitch angle should be 10.45 deg and the rotor speed should be 12.1 rpm for the NREL 5-MW baseline wind turbine.
Best regards,
Dear Jonkman,
Thank you for your reply.
I tried to debug the test 18 but I am still facing the same issue.
I have attached the other parameters of test 18 for the conditions given below to receive your opinion on to where I am possibly going wrong.
Steady wind speed is15 m/s, the blade-pitch angle is at 10.45 deg. There is an unusual increase in the rotor speed from somewhere around 12 rpm to around 18.5 rpm over the time span of 60 seconds.
Moreover, PCMode and VSContrl are 0 in the ServoDyn file.
Best Regards,
Kamran Ali
Dear Kamran,
To get a rotor speed of 12.1 rpm for the NREL 5-MW baseline wind turbine at a steady wind speed of 15 m/s and a steady blade-pitch angle of 10.45 deg, you must have a constant generator torque of 43093.55 Nm (i.e. rated torque). You’ve set VSContrl = 0, which implies that ServoDyn will use a generator model (GenModel), which I presume you have not set to yield constant torque based on your resulting generator power and torque. To set constant torque in ServoDyn, it is easiest to set VSContrl = 1, VS_RtTq = 43093.55 Nm (or whatever constant torque you want) and VS_RtGnSp = VS_Rgn2K = VS_SlPc = some small number greater than zero e.g. 0.001.
I hope that helps.
Best regards,
Dear Jonkman,
Thank you for pointing the issue out for me.
You were right, I had not set the torque constant and I was able to get the desired results keeping torque constant. however, there is slight fluctuation in the rotor speed during the first 10s of the simulation, I am wondering if it’s a normal behavior or not?
Best Regards,
Kamran
Dear Kamran,
From your plot, it looks like you’ve initialized the rotor speed to 12.1 rpm, but the steady-state rotational speed is closer to 11.81 rpm for the conditions you’ve set. The start-up transient is simply the initial condition solution of the model decaying away until the steady-state condition is reached.
Best regards,
