I am using FAST coupled to a bladed controller DLL. Now I want to do simulations with an pitch offset of 0° / 0.3° / -0.3° for the three blades, i.e. these values should be added by FAST to the values returned from the controller DLL to simulate some aerodynamic asymmetry or blade mounting errors. Is there a way to achieve this with FAST?
In the current version of FAST, the only way to model a pitch imbalance is by introducing pitch angle differences between the blades. FAST allows independent blade-pitch control, so, pitch imbalances are possible in FAST if the controlled pitch angles include the pitch error. If your Bladed DLL does not already include the pitch imbalance in the pitch output, what you can do is modify the PitchCntrl() routine in source file BladedDLLInterface.f90 to add the imbalance to the pitch commands before returning. This is not the most ideal solution, but it is simple and will work. We have on our to-do list a plan to add pitch imbalance inputs to the FAST input file, but we simply haven’t gotten around to it yet.
Thank you for your reply. I already realized that introducing pitch imbalance is possible by modifying the source code of the DLL-Interface. However this is kind of ugly, so I hope, the next version of FAST will include an option in the FAST input file.
I’m studying the effects of aerodynamic rotor imbalance on power production and fatigue of wind turbine components. One case I’m considering is a constant pitch offset affecting only one blade.
To do so, I am using the 5-MW Reference Offshore Wind Turbine and, to create a constant pitch offset, I’m adding an identical value of error to every point of the BlTwist column in the “NRELOffshrBsline5MW_AeroDyn_blade.dat” of the faulty blade. Is this approach correct?
During my simulations, I noticed something that caught my attention. When I apply negative values for the offset, the results (please, see the attached figures) in steady state are different depending on the faulty blade. However, when I apply positive pitch offsets, the results are basically the same, independent of the faulty blade. Could someone help me understand why this is happening?
I’m also attaching the log files for the simulations, just in case you might want to take a look at the warnings that appeared. I’m using constant wind speed (11 m/s) and Elastodyn for the blades. I’m assuming that the wind turbine is installed onshore, similar to the 5MW_Land_DLL_WTurb test case, but with constant uniform wind.
Pitch Offset - Log Files.zip (13.3 KB)
Yes, adding a constant offset to BlTwist will result in an aerodynamic pitch imbalance (this feature was not available when I responded to Felix in Dec, 2010).
What is the difference between B1/-2, B2-2, and B3-2 (likewise for +)? Are you simply toggling which blade has the twist offset / imbalance? Due to the symmetry, I would expect identical results for power, thrust, etc. when toggling which blade has the imbalance.
I see that you have strong start-up transients over the first 70-s of simulation; this likely implies the setting of poor initial conditions, which may cause lingering problems in your simulation.
I also see something odd in your log files: every node of every blade is experiencing high angle of attack errors; I would not expect this error (except at the nodes nearest the root) during normal operation.
I would guess something is set up incorrectly in your model, but it is hard for me to guess without knowing more.
Thank you for the reply.
I’m sorry for not making the legend clearer. The idea is:
- B1/-2°: Blade 1 has a pitch offset of -2°. The other blades have no offset.
- B2/-2°: Blade 2 has a pitch offset of -2°. The other blades have no offset.
- I used the same logic for B3/-2° and for the +2° cases.
Now, regarding the initial conditions, I’ve set them all to zero in the “NRELOffshrBsline5MW_Onshore_ElastoDyn.dat” file.
I will attach the files I’m using to run the simulations. “NRELOffshrBsline5MW_AeroDyn_blade3+2.dat” corresponds to the case where blade 3 has a +2° pitch offset and “NRELOffshrBsline5MW_AeroDyn_blade3-2.dat” to the -2° pitch offset case.
I’m not attaching the files that are called from the “5MW_Baseline” folder. The ones I have changed were put in the same directory of the .fst file.
Could you help me find out if something is out of the ordinary?
Pitch Offset - Files for simulation.zip (25.4 KB)
From my brief skim of your input files, I don’t see any obvious problems other than the poor initial conditions. Instead of setting all initial conditions to zero, it is generally recommend to set the initial rotor speed and initial blade-pitch angles in ElastoDyn to be their expected values for the mean wind speed you are using. For the NREL 5-MW baseline turbine at 11 m/s, this would be RotSpeed = 11.89 rpm and BlPitch = 0.0 deg. Does using these initial conditions result in the response you expect?
Thank you for the reply.
I’ve run the simulations using the initial conditions you suggested and I noticed that the warnings due high angles of attack were not displayed. Also, the stars-up transients seem to have been softened.
After taking a closer look at my python routines, I’ve noticed that during the manipulation of the “NRELOffshrBsline5MW_AeroDyn_blade.dat” files an error was being introduced in the negative pitch offset cases. Now the results don’t depend on the faulty blade.
Thank you again for your help.
Pitch Offset - Log Files - New.zip (5.15 KB)
I’m glad the problem is solved!
I am trying to add an equal pitch angle offset for every blade of a DTU 10MW RWT. However, since I didn’t find any aerodyn v.15 inputfiles for this model, I am still working with aerodyn v.14 inputfiles. In the blade inputfile I have a column for StrcTwst instead of BlTwist. Does adding a constant angle to StrcTwst have the same effect as adding a constant angle to BlTwist?
Thank you in advance!
In AeroDyn v14, the blade aerodynamic twist is specified via AeroTwist (the second column of the table). All blades in AeroDyn v14 must have the same aerodynamic properties (this table is not defined uniquely for each blade).
The structural twist (StrcTwst) in ElastoDyn defines the orientation of the principle axes of bending, not the aerodynamic twist.
Dear Dr. Jason Jonkman,
Thank you very much for giving me this insight.