Replies: 2 comments 1 reply
-
Dear @Laurens09, Ignoring the details of your simulation set-up at first (which I haven't reviewed yet), it doesn't appear that your solution has converged and reached a periodic steady state yet. You appear to be showing only results over the first 1.5 seconds of a longer simulation. Does the solution reach a periodic steady state if you run the simulation for longer? Best regards, |
Beta Was this translation helpful? Give feedback.
-
Dear @Laurens09, The load case that you are running seems to be the Load Case 2.16 from OC6 phase III (https://wes.copernicus.org/articles/8/465/2023/wes-8-465-2023-discussion.html). I think the only difference is that you are running with the original UNAFLOW definition of 39 aerodynamic sections instead of the simplified 20 sections that we used in the OC6 phase III project. Below you can see the response in time domain that we obtained when using OLAF: Somehow, your plots are very noisy. This is not expected for a spatially uniform wind inflow. By the way, I believe both plots that you shared correspond to aerodynamic torque. I suspect the RotMotionFile could be wrong. Note that you should provide 4 columns with:
And the azimuth must be an unwrapped angle. See below for reference: I would probably recommend starting without platform motion nor rotor speed variations and check the outputs against Figures 3 and 4 from the OC6 phase III paper commented above. I hope that helps! |
Beta Was this translation helpful? Give feedback.
-
Dear all,
I have created a FVW model of a wind turbine through the OLAF module of openfast. The WT platform oscillates in the surge direction (frequency = 2Hz) and the rotor speed oscillates also (frequency = 2Hz). The difference of phase between the 2 oscillations is 180°.
The solution given by the model doesn't converge to the rigth one (see figures below: X-axis = Time [s], Y-axis = Torque [Nm] or Thrust [N] ). I would expect that thrust and torque oscillate with a frequency of 2Hz.
I have tried several settings (DT, number of panels, CircSolvConvCrit, CircSolvRelaxation, CircSolvMaxIter...), but I'm not able to get good results. Do you have any suggestions to solve the problem?
Thank you
AERODYN.dat, Aerodyn_driver.dvr, OLAF.dat are attached below.
*The red line is the mean expect value.
AERODYN.dat
Aerodyn_driver.dvr
OLAF.dat
Beta Was this translation helpful? Give feedback.
All reactions