Dear all,
I’m trying to simulate a case of a wind turbine (5MW_OC3Mnpl) yawing with the uniform wind (WindType=2) direction using the latest OpenFAST. In the uniform wind file, the WindDir is set to 0, -90 and -180 (deg) at the Time of 50s, 140s and 230s (wind direction changes -1deg per second). In ServoDyn file, TYawManS=50, YawManRat=1 and NacYawF=180. YawDOF=True in ElastoDyn file. But FAST aborted when YawPzn reaches about 80deg with the following error. This will occur even the wind speed is small.
I can’t find the reason of this error. Any help will be appreciable.
Best regards.
Dear @Menghao.Yao,
Can you confirm that the wind direction and nacelle-yaw angle outputs are linearly varying as you specified?
I see the small angle assumption violation warning, which suggests your model has some issues, but it is not clear to me from what you’ve shared if this is caused by the high skew angle or if the high skew angle is caused by the large rotation.
Best regards,
Dear @Jason.Jonkman,
Thanks for your quick reply. The following is the wind direction and nacelle-yaw angle outputs. They changed exactly as I specified.
And I can simulate this case using FAST7.
Best regards,
Dear @Menghao.Yao,
Then I would conclude that the high skew angle is caused by the small angle assumption violation error triggered by the large tower deflection. My guess is that your model is unstable for some reason. You mention that you are using the model of the NREL 5-MW baseline turbine atop the OC3-monopile. Did you change the model provided by NREL in any way, e.g. increasing the time step or increasing a stiffness?
Best regards,
Dear @Jason.Jonkman,
I’ve rechecked my model settings and confirmed that nothing has changed to the model except the wind and nacelle-yaw setting mentioned above. Then I reduced the time step to 0.0025 and ran the simulation, FAST aborted when YawPzn reaches about only 55deg with the same error.
Best regards,
Dear @Menghao.Yao,
Are you using the OpenFAST model of the NREL 5-MW baseline wind turbine atop the OC3-monopile from the OpenFAST r-test: r-test/glue-codes/openfast/5MW_OC3Mnpl_DLL_WTurb_WavesIrr at main · OpenFAST/r-test · GitHub?
It sounds like all you’ve changed is the wind data file and the yaw maneuver. Can you share these updated InflowWind and ServoDyn inputs?
Best regards,
Dear @Jason.Jonkman,
Yes, I’m using the model from the OpenFAST r-test on GitHub. I’ve uploaded the changed files at this link: Microsoft OneDrive - Access files anywhere. Create docs with free Office Online.. The wave load is also turned off by setting WaveMod=0.
Best regards,
Dear @Menghao.Yao,
I was able to reproduce the error you are facing, which seems to be some numerical solver instability. Adding one correction step (NumCrctn = 1 in the OpenFAST primary input file) seems to solve the issue.
Best regards,
Dear @Jason.Jonkman,
Thank you very much for the solution, now I can successfully simulate this case.
By the way, I’d like to mention another problem I encountered while trying to solve the above question. The settings are the same as the above case except for: (1) WindDir in uniform wind file is 20deg with no change. (2) NacYaw in ElastoDyn file is 20deg. (3) Nacelle yaw setting in ServoDyn is the same as OC3Mnpl in OpenFAST r-test (TYawManS = 9999). (4) NumCrctn = 1 as you suggested. Then the following ERROR occurred. I thought the response should be the same as when the wind direction and nacelle-yaw angle are 0. I found that it seems that FAST will abort when initial nacelle-yaw angle is larger than 18deg if YawDOF = true. Is this normal?
Best regards,
Dear @Menghao.Yao,
Have you set both the initial nacelle-yaw angle in ElastoDyn (NacYaw) and the neutral yaw position in ServoDyn (YawNeut) to the same value? If not, this will cause the nacelle-yaw angle to quickly snap back from NacYaw to YawNeut and result in unrealistic loads and deflections as you are showing. If you want to keep the nacelle at 20deg and keep the nacelle-yaw DOF enabled (YawDOF = TRUE), then set YawNeut = 20deg as well.
Best regards,
Dear @Jason.Jonkman,
Thank you for your answer, it really helps.
Beast regards,
