Dear all,
After performing simulation (with turbsim generated wind file *.bts as input) in FAST, I’ve extracted the wind velocities (Vx, Vy, Vz) from the .out file generated by FAST. Is this rotationally sampled wind ? or Eulerian wind measurement (i.e wind measured at an inertial location) ? My understanding is, it is the later (Eulerian wind measurement) ?
Regards,
Kumara
Dear Kumara,
You haven’t stated which FAST outputs you are using. In general, wind output at fixed points would be in the inertial frame and wind output in a frame rotating with the rotor would be rotationally sampled. For example, output Wind1VelX from InflowWind would be in the inertial frame and output B1N1VUndx from AeroDyn v15 would be rotationally sampled.
Best regards,
Dear Jason,
I was talking about “Wind1VelX”. Thanks for the clarification.
Regards,
Kumara
Dear @Jason.Jonkman,
Does OpenFast use a fixed reference turbulence or a rotationally sampled wind turbulence as the “InflowWind” input?
In other words, do I need to generate a fixed reference turbulence profile (using TurbSim, for example) and provide it as an input file to OpenFast, and then OpenFast will internally apply rotational sampling when necessary? Or can I directly generate a rotationally sampled wind profile (using a mathematical model of the turbulence with the necessary trigonometric transformations) and provide it as input to OpenFast?
Best Regards.
Dear @Bilal.Tout,
OpenFAST uses wind turbulence as input specified relative to a fixed frame of reference and the rotationally sampled wind field is processed by the spinning blades,
Best regards,
Dear @Jason.Jonkman,
Thank you for your time and assistance.
How can I visualize the processed wind field (rotationally sampled)? I tried to extract the B1N1VUndx file from AeroDyn v15, but it returns a zero signal.
You can find the input files, as well as the output and summary files for my simulation, via this link: InputWind_test - Google Drive .
Best Regards.
Dear @Bilal.Tout,
From my quick look of your AeroDyn input file, I see that you have not defined blade nodal outputs, which is why these outputs are not available. If you want to select blade nodes for outputs, set NBlOuts > 0 and set BlOutNd to a list NBlOuts-long of the aerodynamic analysis nodes you want to output.
Best regards,
Dear @Jason.Jonkman,
Thank you, this works well.
I have another question regarding the wind input to the wind turbine linearized model, if the following procedure is used to obtain the linearized LTI model:
Does the linearized model take also the fixed wind profile as the disturbance input? Or, should I use the equivalent wind model (with admittance filters that allows rotational sampling) in this case?
Best Regards.
Dear @Bilal.Tout,
Good question. The LTI model will not rotationally sample the inflow perturbations. That said, rotational sampling is important for full-field turbulence, whereas the LTI model is only valid for uniform inflow. Perhaps the effect of rotational sampling could be considered as you describe, but I have not pursued such an approach myself.
Best regards,
Dear @Jason.Jonkman ,
Does this apply whatever wind type is being used (WindType = 2 : uniform, WindType = 3: Binary .bts)? i.e., if i generate a wind profile using an approximation of the kaimal spectrum and store values in a (.dat) file then use it as input to OpenFast using uniform wind input (WindType = 2), does Openfast apply internally the rotational sampling? Or is it only applied when using a full-field wind generated using TurbSim?
Thanks in advance.
Regards.
Dear @Bilal.Tout,
Yes, any inflow field used within OpenFAST will be rotationally sampled, but keep in mind that WindType = 2 is time-varying, but spatially uniform (other than shear), so, there is little rotational sampling to be had with WindType = 2.
Best regards,