Good morning all,
I am trying to simulate one of the sub cases in DLC 22 for the NREL 5MW according to the DTU Design Load Cases definition, which is DLC22y - Power production with abnormal yaw error. The simulation parameters are 4-26 m/s in steps of 2 m/s, 1 turbulence seed with NTM model per wind speed and a yaw angle range of 15 deg to 345 deg in steps of 15 deg. In total there are 276 simulation files. However some of the simulations fail and the error message is shown below -
FAST_Solution:FAST_AdvanceStates:SolveOption2c_Inp2AD_SrvD:InflowWind_CalcOutput:CalcOutput:IfW_Bl
adedFFWind:CalcOutput [position=(-45.747, -29.214, 7.9952)]: FF wind array boundaries violated.
Grid too small in Z direction (height (Z=7.9952 m) is below the grid and no tower points are
defined).
FAST encountered an error at simulation time 93.844 of 700 seconds.
Simulation error level: FATAL ERROR
Aborting OpenFAST.
The TurbSim parameters I have used to create these wnd files are -
--------Turbine/Model Specifications-----------------------
31 NumGrid_Z - Vertical grid-point matrix dimension
31 NumGrid_Y - Horizontal grid-point matrix dimension
0.05 TimeStep - Time step [seconds]
700 AnalysisTime - Length of analysis time series [seconds] (program will add time if necessary: AnalysisTime = MAX(AnalysisTime, UsableTime+GridWidth/MeanHHWS) )
“ALL” UsableTime - Usable length of output time series [seconds] (program will add GridWidth/MeanHHWS seconds unless UsableTime is “ALL”)
90.0 HubHt - Hub height [m] (should be > 0.5GridHeight)
170.0 GridHeight - Grid height [m]
180.0 GridWidth - Grid width [m] (should be >= 2(RotorRadius+ShaftLength))
0 VFlowAng - Vertical mean flow (uptilt) angle [degrees]
0 HFlowAng - Horizontal mean flow (skew) angle [degrees]
In all the failed cases, the error message is the same, that the grid is too small in Z direction. Most of the failed cases have a high nacelle yaw angle (~100 deg). My question is, if the grid has the correct dimensions for the cases that ran successfully, how is it that turbine yaw causes the grid to become small in the Z direction for the failed cases? The turbine dimensions have not changed in the Z direction, just the nacelle has been given different yaw angles for each case.
It would be very helpful to get some guidance in this matter. Apologies if I have posted this in the wrong place.
Thanks in advance!
Best regards,
Vishal
Dear Vishal,
I’m surprised you’d receive this error if you are using the TurbSim specifications shown. The TurbSim specifications show that the wind field grid extends from Z = 5- to 175-m above the ground, and the error is stating that Z = 7.9952 m is below the grid. Are you sure you are using the correct TurbSim wind data file?
Regardless, I would check the wind turbine output parameters make physical sense for the simulation set up you are using. Are, e.g., the nacelle-yaw angle, rotor speed, blade-pitch, and blade and tower deflection outputs reasonable for the simulation(s) your running?
Best regards,
Dear Jason,
Apologies, I posted some old TurbSim parameters I was using, I got the error message in the last post when I used the following TurbSim parameters -
--------Turbine/Model Specifications-----------------------
31 NumGrid_Z - Vertical grid-point matrix dimension
31 NumGrid_Y - Horizontal grid-point matrix dimension
0.05 TimeStep - Time step [seconds]
700 AnalysisTime - Length of analysis time series [seconds] (program will add time if necessary: AnalysisTime = MAX(AnalysisTime, UsableTime+GridWidth/MeanHHWS) )
“ALL” UsableTime - Usable length of output time series [seconds] (program will add GridWidth/MeanHHWS seconds unless UsableTime is “ALL”)
120.0 HubHt - Hub height [m] (should be > 0.5GridHeight)
200.0 GridHeight - Grid height [m]
180.0 GridWidth - Grid width [m] (should be >= 2(RotorRadius+ShaftLength))
0 VFlowAng - Vertical mean flow (uptilt) angle [degrees]
0 HFlowAng - Horizontal mean flow (skew) angle [degrees]
When I use these parameters, the wind field grid extends from 10 m to 210 m above ground, and the error message that Z = 7.9952m is in fact below the grid. I am using the correct .wnd files, and for this simulation, the mean wind speed is 26m/s (however when the wind field is loaded, it says - Reading a 31x31 grid (180 m wide, 10 m to 210 m above ground) with a characteristic wind speed of 27.54 m/s.). Could this cause any problem?
I have looked at the results until the simulation failed, and the operational parameters are completely non-physical. I have attached plots of RPM, Pitch and Nacelle Yaw below -
As you can see, the nacelle yaw is 105 deg for this simulation, but the RPM and pitch values are non-physical. What could this be causing this? Every other parameter is the same for all 276 simulations except the Nacelle yaw, initial RPM and pitch settings, and the .wnd files for every wind speed.
Best regards,
Vishal
Dear Vishal,
It looks like the wind turbine controller has gone unstable, as evident by the large oscillation in blade-pitch angle and lack of rotor-speed regulation.
I would guess that this could be caused by incorrect initial conditions in the controller. Because you are simulating with large yaw error, I suggest using the wind speed in line with the shaft as the wind speed used to set the initial rotor speed and blade-pitch angle. That is, instead of using Vhub, I would would use Vhub*COS(yaw error) as the wind speed used to set the initial rotor speed and blade-pitch angle.
Best regards,
Dear Jason,
I reset the initial conditions according to VHub*Cos(yaw), and this solved the problem for quite a few of the simulations, they ran completely fine. But there were still some that failed, and due to the same reason (grid too small in Z direction). I changed the turbulence seed for the ones that failed and some of the failed ones ran correctly. What I saw in the time history of wind speed in the flow direction was that the maximum value went in excess of 32 m/s for a mean wind speed of 24 m/s. I felt that this is too large a deviation, and maybe that is causing instabilities in the simulations that failed. Do you think that can cause any issues?
Best regards,
Vishal
Dear Vishal,
I’m glad you solved the main problem.
Really high wind speeds result in large pitch angles, which can result in abnormal blade deflection, e.g., deflection upwind. But I’d have to understand more about the cases that fail to really understand why some still result in an instability.
Best regards,
Dear Jason,
I am attaching a pdf with the input files I have used for one of the cases which ran completely but does not have meaningful results. The case in question is for a mean wind speed of 24 m/s, Normal Turbulence Model, -165 deg nacelle yaw. I have set the initial RPM and pitch values in the ElastoDyn input file corresponding to a wind speed of 22 m/s (since the simulation failed very early on for initial settings corresponding to a wind speed of 24*Cos(165) ~= 23 m/s).
I am also attaching plots of RPM and pitch time series for this simulation, and as you can see, the RPM goes unstable early on in the simulation but it runs for 700 seconds. It would be very helpful to get your insight on this.
Thanks in advance,
Best regards,
Vishal
Input Files.pdf (151 KB)
Dear Vishal,
It appears that the controller is going unstable. In reality, 24*COS(165) = -23 m/s, meaning, the flow is passing through the rotor backwards. The controller is not designed for this condition, so, I’m not surprised that the controller is unstable. Perhaps there are initial conditions that could result in a stable control response, but I’m not sure what those would be. I’m not really sure why you are running an operational case with such a large yaw error.
Best regards,
Dear Jason,
I am simulating many load cases specified by the IEC standard for onshore turbines and I am using a reduced Design Load Basis document used by DTU. One of the load cases specifies power production with failure of yaw system leading to abnormal yaw error, in the operating range of 4-26 m/s with Normal Turbulence Model and yaw errors of 15 to 345 degrees in steps of 15. The case that I mentioned occurred for this DLC.
I will re-evaluate my design specifications and the conditions I have to analyze, but as an initial run, I thought I would perform the entire Design Load Basis.
Thank you for your inputs!
Best regards,
Vishal
Dear Vishal,
I’m familiar with the loads analysis requirements of IEC 61400-1 (wind turbines), -3-1 (fixed offshore), and -3-2 (floating offshore), and these design standards do not require simulations of an operational wind turbine at such high yaw errors. (The extreme coherent gust with direction change (ECD) load case is probably the closest.) I’m not familiar with the DTU document you are referring too, but I would guess the supervisory system would kick in and shut down most turbines during such a condition.
Best regards,
Hi jason
are you familiar with loads analysis requirements of IEC 61400-2 (small wind turbine)?
i have a table for IEC61400-1 (onshore wind turbine) from DTU (image bellow) and want to have similar file for small wind turbine
Dear @Ali.Rouhbakhsh,
There is a load case table in IEC 61400-2, that has many similarities to the equivalent table from IEC 61400-1, but I can’t share the IEC document here due to copyright protection.
Best regards,
Yes, I have seen the load case table in IEC-61400-2
Please pay attention to the image below and the parameters that I have circled in red
I don’t know what value to set this parameter for IEC61400-2?
(i want to have a similar table for load case of IEC61400-2)
Dear @Ali.Rouhbakhsh,
These details are discussed in the -2 design standard and will change depending on which edition of the standard you are referencing.
Best regards
But I did not find this detailس in 61400-2
For example, I didn’t see anywhere about the number of seeds
Dear @Ali.Rouhbakhsh,
I’m not an expert on the -2, but the lack of specifics likely implies that you’ll have to run enough seeds until convergence is reached. You may be able to justify using the same number of seeds as called out in the -1 standard. If you are working with a certification body, you could ask them for guidance as well.
Best regards,
Dear @Jason.Jonkman
Currently, I am running simulations using the IEA 15MW UMaine semi-submersible model according to IEC 61400-3-1,-2 (ed. 1.0) and found some DLC 6.2 and DLC 7.1 time series data have large oscillations.
I set the initial parked condition according to 4.2.1.5. Modeling Considerations — OpenFAST v3.5.2 documentation
- DLC 6.2 condition = Wind speed: 47.5 m/s, Wind direction: 0 (fixed), Wind-wave misalignment: -90:30:90, Nacelle Yaw: -165:15:180
- Yaw misalignment angle where large oscillations shown : ±15, ±30, ±45, ±60, ±135
- DLC 7.1 condition = Wind speed: 38.0 m/s, Wind direction: 0 (fixed), MIS: -90:30:90, Nacelle Yaw: -170:10:180
- Yaw misalignment angle where large oscillations shown : ±10, ±20, ±30, ±40, ±50, ±60, ±130, ±140
Some particular yaw misalignment angle causes a large oscillation.
You said the controller and initial condition might cause oscillation, but the parked condition doesn’t use the controller.
Would you please give me some advice for this situation?
Best regards,
Sangwon
- Example of DLC 6.2 abnormal time series data
- Corresponding DLC 6.2 OpenFAST input parameter (fst, AeroV15Dyn, ElastoDyn, ServoDyn)
- Example of DLC 7.1 abnormal time series data
- Corresponding DLC 7.1 OpenFAST input parameter (fst, AeroV15Dyn, ElastoDyn, ServoDyn)
Dear @Sangwon.Lee,
At certain yaw errors in the idling condition, a known blade-edgewise instability exists, e.g., as discussed in the following forum topic: Designing for yaw errors using FAST.
Best regards,
1 Like
Dear @Jason.Jonkman
Thank you for your information.
It was helpful and gave me a clue.
Have a nice day!
Best regards,
Sangwon
1 Like
