Hi all,
somehow Turbsim will show me a severe error while running with User time series. Everything workes normally when nComp in the User Time Series file is set to 3 with a constant w=0.
Unfortunately, as soon as I set nComp to 2 (since I only have u and v components in my input file), TurbSim tells me the followiing after startup:
Running TurbSim (v2.00.07a-bjj, 14-Jun-2016).
Reading the input file "TurbSim.inp".
Reading the user-defined time-series input file ".\Messpunkt01.inp".
forrtl: severe (157): Program Exception - access violation
Image PC Routine Line Source
TurbSim_x64.exe 00007FF7817EEB5B Unknown Unknown Unknown
TurbSim_x64.exe 00007FF781709AA3 Unknown Unknown Unknown
TurbSim_x64.exe 00007FF7817409D5 Unknown Unknown Unknown
TurbSim_x64.exe 00007FF781D9D2E6 Unknown Unknown Unknown
TurbSim_x64.exe 00007FF7822C7E9C Unknown Unknown Unknown
KERNEL32.DLL 00007FFD3B6A4034 Unknown Unknown Unknown
ntdll.dll 00007FFD3DF63691 Unknown Unknown Unknown
Is there a way to solve this problem? With nComp=3 the programm is working quiet well, but since in real life w-velocity values will never be zero for a longer period, it would probably be better simulating without a w-value in the input file.
Thank you guys so much in advance and
greetings from Germany
Jan Spitzer
Hi, Jan.
I think this bug has been fixed in the latest version of OpenFAST (github.com/openfast/openfast). It might be in the master branch, but I’m pretty certain it was merged into the dev branch. Let me know if that doesn’t fix the problem.
When you set nComp to 2, though, it just sets w to 0.
I never had the time/funding to add a method to estimate the full-field turbulence for unmeasured velocity components while I was at NREL. You could probably use the user-defined spectral model to generate a somewhat realistic vertical component based on the longitudinal/lateral wind-speed measurements (or just use the standard Kaimal model), and then combine that simulated vertical wind speed with the longitudinal and lateral wind speed simulated with the user time series method. It’s a lot of extra work, though.
Hi Bonnie,
thank you so much for your quick answer!
Especially the reminder of TurbSim setting the w-component to zero automatically helped a lot.
Best regards
Jan
Hi all,
I want to generate a binary full-field turbsim file (.bts), WindType = 3 in Inflowfile. Is there a way to generate .bts file such that the effect of turbulence only in the X-direction is considered and the velocity in the y and z direction will be zero?
Thanks,
Shubham
Dear @Shubham.Baisthakur,
I have not done this myself, but TurbSim should be able to generate such a file. One way I can think would work would be to select TurbModel = “USRINP” and then in the UserFile, set the spectra (or the spectra scaling) of the v- and w-components to zero. I would guess that you’d also want to set SCMod2 = SCMod3 = “NONE” to avoid computing coherence in the v and w directions as well.
Best regards,
Hello Jason,
Thanks for your reply. Can you please tell me which version of Turbsim you are referring to and where I can find it? In the Turbsim input file I am using, there is no option to specify the “USRINP” as the turbulence model. Also, it does not have the parameters SCMod2 and SCMod3.
Thanks,
Shubham
Dear @Shubham.Baisthakur,
I’m referring to the current version of TurbSim supplied with OpenFAST on github: openfast/modules/turbsim at main · OpenFAST/openfast · GitHub. You could also use TurbSim v2.00.07a-bjj available from the Wind Data and Tools page: TurbSim | Wind Research | NREL.
Best regards,
Thanks @Jason.Jonkman. I will try this method.
I was running IEA-15MW wind turbine simulations for a steady wind speed of 7m/s. I am including the plots of undisturbed wind velocity in the Y and Z directions at the blade nodes.
Does the velocity component in the Y and Z directions only come from the inclination of the blade local coordinate system with the inertial coordinate system, or is there some other factor involved?
Thanks,
Shubham
Dear @Shubham.Baisthakur,
Yes, if you are using WindType = 1 (steady wind) in InflowWind with PropagationDir = VFlowAng = 0, then the inflow is only along global X axis of the inertial frame coordinate system. So, y and z values in the local blade coordinate system are purely the result of the orientation of the local blade coordinate system relative to the global inertial frame.
Best regards,
Hello @Jason.Jonkman
Referring to the following image from the Aerodyn manual, is it correct to assume that the undisturbed wind velocity in the X-direction acts along the local y-axis while the undisturbed wind velocity in the y-direction acts along the local x-axis?
Dear @Shubham.Baisthakur,
No. If the turbine has no yaw, tilt, precone, or elastic deflections, and for a blade pointed vertically upward, the local blade coordinate system would be aligned with the global inertial frame coordinate system with x=X, y=Y, and z=Z. But any nonzero yaw, tilt, precone, elastic deflection, or azimuth angle would cause a deviation between these coordinate systems.
Best regards,
