NcIMUT is very large value

[Hong.Wu]

Hi forum,

I am using OpenFAST to simulate a floating wind turbine platform with the IEA15MW wind turbine and a floating tower. The wind speed is 49 m/s, the wave mode is JONSWAP, with Hs = 8m and Tp = 13.9s. I have disabled SeverDyn, and the pitch angle is set to 90°. Currently, I am seeing very large values for the nacelle acceleration, as shown in the figure below. Could this indicate tower instability?

OpenFAST-v4.0.3-2-g5f4d8a79
Compile Info:
 - Compiler: Intel(R) Fortran Compiler 20250100
 - Architecture: 64 bit
 - Precision: single
 - OpenMP: No
 - Date: Apr 10 2025
 - Time: 10:24:39
Execution Info:
 - Date: 09/06/2025
 - Time: 00:03:01+0800

OpenFAST input file heading:
    -- IEA 15 MW offshore reference model on SemiSpar floating platform

 Running ElastoDyn.
 Running InflowWind.
 Running SeaState.
 Generating incident wave kinematics and current time history.
 Running AeroDyn.
   AeroDyn: projMod: 1
   Projection: legacy (NoSweepPitchTwist), BEM: legacy (2D)
 Running HydroDyn.
 Reading in WAMIT output with root name "../SemiSpar-LG_Parametric/../HydroData/semiSparLG".
 Computing radiation impulse response functions and wave diffraction forces.
 Running MoorDyn (v2.2.2, 2024-01-16).
   This is MoorDyn v2, with significant input file changes from v1.

**************************************************************************************************
MoorDyn

Copyright (C) 2025 National Renewable Energy Laboratory
Copyright (C) 2025 Envision Energy USA LTD

This program is licensed under Apache License Version 2.0 and comes with ABSOLUTELY NO WARRANTY.
See the "LICENSE" file distributed with this software for details.
**************************************************************************************************

   Parsing MoorDyn input file: ../SemiSpar-LG_Parametric/IEA-15-RWT-SemiSpar_MoorDyn.dat
   Created mooring system:  18 lines, 24 points, 0 rods, 0 bodies.
   Finalizing initial conditions using dynamic relaxation.
   t=0.34  FairTen 1: 2.02593E+06, 2.02616E+06, 2.02612E+06
   Fairlead tensions converged to 0.1% after 0.34 seconds.
   Fairlead tension: 2.02593E+06
   Fairlead forces: 1.91176E+06, -3.37085E+05, -5.79620E+05
   Fairlead tension: 2.02721E+06
   Fairlead forces: 1.90988E+06, -3.36777E+05, -5.90374E+05
   Fairlead tension: 2.05173E+06
   Fairlead forces: 1.90984E+06, -3.36772E+05, -6.69844E+05
   Fairlead tension: 2.02593E+06
   Fairlead forces: 1.91176E+06, 3.37085E+05, -5.79620E+05
   Fairlead tension: 2.02721E+06
   Fairlead forces: 1.90988E+06, 3.36777E+05, -5.90374E+05
   Fairlead tension: 2.05173E+06
   Fairlead forces: 1.90984E+06, 3.36772E+05, -6.69844E+05
   Fairlead tension: 2.02577E+06
   Fairlead forces: -6.63881E+05, 1.82400E+06, -5.79664E+05
   Fairlead tension: 2.02713E+06
   Fairlead forces: -6.63271E+05, 1.82232E+06, -5.90346E+05
   Fairlead tension: 2.05166E+06
   Fairlead forces: -6.63256E+05, 1.82228E+06, -6.69835E+05
   Fairlead tension: 2.02614E+06
   Fairlead forces: -1.24794E+06, 1.48726E+06, -5.79615E+05
   Fairlead tension: 2.02712E+06
   Fairlead forces: -1.24655E+06, 1.48560E+06, -5.90167E+05
   Fairlead tension: 2.05163E+06
   Fairlead forces: -1.24643E+06, 1.48546E+06, -6.70062E+05
   Fairlead tension: 2.02614E+06
   Fairlead forces: -1.24794E+06, -1.48726E+06, -5.79615E+05
   Fairlead tension: 2.02712E+06
   Fairlead forces: -1.24655E+06, -1.48560E+06, -5.90167E+05
   Fairlead tension: 2.05163E+06
   Fairlead forces: -1.24643E+06, -1.48546E+06, -6.70062E+05
   Fairlead tension: 2.02577E+06
   Fairlead forces: -6.63881E+05, -1.82400E+06, -5.79664E+05
   Fairlead tension: 2.02713E+06
   Fairlead forces: -6.63271E+05, -1.82232E+06, -5.90346E+05
   Fairlead tension: 2.05166E+06
   Fairlead forces: -6.63256E+05, -1.82228E+06, -6.69835E+05
   MoorDyn initialization completed.

FAST_InitializeAll:ED_Init:ED_ReadInput:ReadBladeInputs:ReadBladeFile:The ElastoDyn Blade file,
../SemiSpar-LG_Parametric/../../IEA-15-240-RWT/IEA-15-240-RWT_ElastoDyn_blade.dat, DISTRIBUTED
BLADE PROPERTIES table contains the PitchAxis column.  This column is unused and will be removed
in future releases

 Time: 0 of 3600 seconds.
The BEM solution is being turned off due to low TSR.  (TSR = 0). This warning will not be


 Total Real Time:       13.931 minutes
 Total CPU Time:        13.853 minutes
 Simulation CPU Time:   13.103 minutes
 Simulated Time:        60 minutes
 Time Ratio (Sim/CPU):  4.579

 OpenFAST terminated normally.

Please let me know what your thoughts are on these errors, and I am happy to provide more information.

Thank you!

[Jason.Jonkman]

Dear @Hong.Wu,

It is taking a while before your model goes unstable, so, I’d be curious what changes before the instability. For example, are you seeing a sizeable inflow-skew error (due to sizeable yaw motion) that could result in the classic blade edgewise instability in deep stall?

Best regards,

[Hong.Wu]

Dear Jason,

I really appreciate your prompt response.

The wind speed is steady, the rotor speed is 0, , and there are irregular waves. The inflow-skew has a small value before the instability occurs. Regarding the platform’s yaw and sway motion, the side-to-side motion of the tower top, the y- and z-direction force on the rotor, they are all fluctuating. However, this is different from the yaw motion instability mentioned in your thesis.

Looking forward your suggestions.

Best regards,

[Jason.Jonkman]

Dear @Hong.Wu,

I don’t see any of these motions going unstable. Which motions are going unstable, commensurate with the nacelle acceleration?

Best regards,

[Riad.Elhamoud]

Hi @Hong.Wu,

I am Riad, a wind turbine enthusiast.

I am not a specialist but maybe the period of the wave is matching that of the pitch motion. And since the tower top acceleration is coupled with platform pitch motion, you see kind of resonance for the tower top acceleration.

Also kindly i ask what the floater you are using and what is the misalignment angle you are considering ?

Best Regards,

Riad

[Hong.Wu]

Dear @Jason.Jonkman ,

You are correct. None of the motions developed instability.

（1）The acceleration at the nacelle in the x-direction is reasonable under the 0 heading wave and wind conditions. However, the accelerations in the y and z directions are very large , which confuses me.

（2）Additionally, when I try to add the second-order wave forces, the calculation throws an error.

OpenFAST

Copyright (C) 2025 National Renewable Energy Laboratory
Copyright (C) 2025 Envision Energy USA LTD

This program is licensed under Apache License Version 2.0 and comes with ABSOLUTELY NO WARRANTY.     
See the "LICENSE" file distributed with this software for details.
**************************************************************************************************   

OpenFAST-v4.0.3-2-g5f4d8a79
Compile Info:
 - Compiler: Intel(R) Fortran Compiler 20250100
 - Architecture: 64 bit
 - Precision: single
 - OpenMP: No
 - Date: Apr 10 2025
 - Time: 10:24:39
Execution Info:
 - Date: 09/09/2025
 - Time: 15:11:21+0800

OpenFAST input file heading:
    -- IEA 15 MW offshore reference model on SemiSpar floating platform

 Running ElastoDyn.
 Running InflowWind.
 Running SeaState.
 Generating incident wave kinematics and current time history.
 Running AeroDyn.
   AeroDyn: projMod: 1
   Projection: legacy (NoSweepPitchTwist), BEM: legacy (2D)
 Running HydroDyn.
 Reading in WAMIT output with root name "../SemiSpar-LG_Parametric/../HydroData/semiSparLG".
 Computing radiation impulse response functions and wave diffraction forces.
 Calculating second order difference-frequency force using the full quadratic transfer function.     
 Calculating second order sum-frequency force using the full quadratic transfer function.
 Running MoorDyn (v2.2.2, 2024-01-16).
   This is MoorDyn v2, with significant input file changes from v1.

**************************************************************************************************   
MoorDyn

Copyright (C) 2025 National Renewable Energy Laboratory
Copyright (C) 2025 Envision Energy USA LTD

This program is licensed under Apache License Version 2.0 and comes with ABSOLUTELY NO WARRANTY.     
See the "LICENSE" file distributed with this software for details.
**************************************************************************************************   

   Parsing MoorDyn input file: ../SemiSpar-LG_Parametric/IEA-15-RWT-SemiSpar_MoorDyn.dat
   Created mooring system:  18 lines, 24 points, 0 rods, 0 bodies.
   Finalizing initial conditions using dynamic relaxation.
   t=0.34  FairTen 1: 2.02593E+06, 2.02616E+06, 2.02612E+06
   Fairlead tensions converged to 0.1% after 0.34 seconds.
   Fairlead tension: 2.02593E+06
   Fairlead forces: 1.91176E+06, -3.37085E+05, -5.79620E+05
   ......
   MoorDyn initialization completed.

FAST_InitializeAll:ED_Init:ED_ReadInput:ReadBladeInputs:ReadBladeFile:The ElastoDyn Blade file,      
../SemiSpar-LG_Parametric/../../IEA-15-240-RWT/IEA-15-240-RWT_ElastoDyn_blade.dat, DISTRIBUTED       
BLADE PROPERTIES table contains the PitchAxis column.  This column is unused and will be removed     
in future releases

 Time: 0 of 3600 seconds.
The BEM solution is being turned off due to low TSR.  (TSR = 0). This warning will not be
repeated though the condition may persist. (See GeomPhi output channel.)

FAST_Solution:FAST_UpdateStates:FAST_AdvanceStates:AD_UpdateStates:BEMT_UpdateStates:UpdatePhi(nod   
e 43, blade 1):BEMT_UnCoupledSolve:There is no valid value of phi for these operating conditions:    
Vx = 2.60498E-02, Vy = -3.6078, rlocal = 104.15, theta = 1.5179, geometric phi = 3.1344. This        
warning will not be repeated though the condition may persist. (See GeomPhi output channel.)
 Time: 400 of 3600 seconds.  Estimated final completion at 15:26:30 (in 0.007 days).

FAST_Solution:FAST_UpdateStates:FAST_AdvanceStates:ED_ABM4:ED_CalcContStateDeriv:SetCoordSy:Small    
angle assumption violated in SUBROUTINE SmllRotTrans() due to a large blade deflection (ElastoDyn    
SetCoordSy). The solution may be inaccurate. Simulation continuing, but future warnings from
SmllRotTrans() will be suppressed.
 Additional debugging message from SUBROUTINE SmllRotTrans(): 404.27 s
ED_CalcContStateDeriv:SetCoordSy:Small angle assumption violated in SUBROUTINE SmllRotTrans() due    
to a large blade deflection (ElastoDyn SetCoordSy). The solution may be inaccurate. Simulation       
continuing, but future warnings from SmllRotTrans() will be suppressed.
 Additional debugging message from SUBROUTINE SmllRotTrans(): 404.27 s
 Time: 770 of 3600 seconds.  Estimated final completion at 15:26:30 (in 0.006 days).
Warning: SkewedWakeCorrection encountered a large value of chi (99.04 deg), so the yaw correction    
will be limited. This warning will not be repeated though the condition may persist. See the AD      
chi output channels, and consider turning off the Pitt/Peters skew model (set SkewMod=1) if this     
condition persists.
 Time: 780 of 3600 seconds.  Estimated final completion at 15:26:30 (in 0.006 
days).

FAST_Solution:FAST_UpdateStates:FAST_AdvanceStates:AD_UpdateStates:BEMT_UpdateStates(node 1,
blade 1):ComputeTau:Uninduced axial relative air speed, Un, is less than 0.1 m/s; limiting
time-varying tau1. This message will not be repeated though the condition may persist.



** WARNING: Tower strike. **  This warning will not be repeated though the condition may persist.    


NaN detected at time 781.59 in Line 3 in MoorDyn.
NaN detected at time 781.59 in Line 3 in MoorDyn.
NaN detected at time 781.59 in Line 3 in MoorDyn.
NaN detected at time 781.59 in Point 2 in MoorDyn.
......

FAST_Solution:FAST_UpdateStates:FAST_AdvanceStates:AD_UpdateStates:SetInputs:SetDisturbedInflow:Tw   
rInfl:Tower strike.
FAST_AdvanceStates: NaN state detected.
 MD_UpdateStates: NaN state detected.


 OpenFAST encountered an error at simulation time 781.6 of 3600 seconds.
 Simulation error level: FATAL ERROR

 Aborting OpenFAST.

Some results as follows:

If you have any ideas, please let me know. I look forward to your reply. Thanks

Best regards,

[Hong.Wu]

Dear @Riad.Elhamoud ,

Thank you for your reply. I am not sure whether it is tower resonance. As I mentioned in my previous message, the acceleration in the x-direction is reasonable, but the values in the y and z directions are very large.

This floater is a new design, with its center of gravity lower than the center of buoyancy, inspired by the spar type.

If you have any ideas, please let me know.

Best regards,

[Riad.Elhamoud]

Dear @Hong.Wu

If were you, i will compute the PSD of tower top acceleration in side-to-side direction and see what it gives.

Maybe you can find an explanation.

Bet Regards,

Riad

[Hong.Wu]

Dear @Riad.Elhamoud ,

That’s a good idea. I performed an FFT by the PyDatView software. The PSD results are as follows. However, what’s frustrating is that I don’t know the natural frequency of the IEA15MW floating tower.

[Riad.Elhamoud]

Dear @Hong.Wu,

I think this maybe explains something or everything.
are yo sure that the x-axis is frequency in Hz or anglar frequency in rad/s.

to know tower natral frequency, i do a free decay test. But maybe you can do it using the Linearization feature in OpenFAST. @Jason.Jonkman do you confirm ? or do yo have an other opinion ?

Best Regards,

Riad

[Hong.Wu]

Dear @Riad.Elhamoud ,

Yes, I think that the x-axis is frequency in Hz. And I performed an FFT for the tower top motion in side-to-side direction, as follows. It also has a peak at 3.7 Hz. Thanks for the method you suggested.

I am curious how this phenomenon is caused — there is no such high frequency in the PSD of my wave lever. Indeed, it seems that some cause (possibly high-frequency loads generated by blade vibration induced by the platform’s base motion) is exciting nacelle acceleration. @Jason.Jonkman What do you think?

In addition, I only performed the floater’s natural frequency from free-decay tests, and I’m not sure how to obtain the tower’s natural frequency by the linearization feature.

Best Regards,

Hong

[Jason.Jonkman]

Dear @Hong.Wu,

For this idling rotor case, I would turn off the wake and unsteady airfoil aerodynamics models of AeroDyn altogether because these models are likely not applicable (by setting Wake_Mod = 0 and UA_Mod = 0). Does this solve the issue?

If not, I would first isolate what mode/degree of freedom is causing the instability. Is this from blade deflection, drivetrain torsion, tower side-side deflection, or platform sway or roll?

You can certainly compute full-system natural frequencies through an OpenFAST linearization analysis (as has been discussed in many other topics on this forum). You could also figure this out through system-ID tests like free decay, but to identify which degrees of freedom correspond to which frequency (blade, drivetrain tower, platform), would likely require that you run separate system-ID tests with different DOFs enabled to isolate each effect. The interpretation is typically more straightforward through a linearization analysis.

Best regards,

[Hong.Wu]

Dear @Jason.Jonkman ,

Thank you sincerely for providing these solutions. I will try to solve the problem using these methods. If issues arise again, I hope I can still get your help.

Beast regards,

[Hong.Wu]

Dear @Jason.Jonkman ,

In the case mentioned previously, I’ve found that there’s a high-frequency component (around 3.7 Hz) first-order wave radiation force in the sway motion (y-axis direction) which causes the force to grow rapidly, eventually exceeding the component in the surge (x-axis) direction. I don’t quite understand where this high-frequency velocity comes from. What are your thoughts on this?

[Jason.Jonkman]

Dear @Hong.Wu,

I would guess the growth in B1RdtFyi is associated with the growth in platform-sway motion and tower side-side motion, not the cause of the motion.

Does this case involve large nacelle-yaw error? I suspect you are experiencing the classic blade-edgewise instability that can happen with large nacelle-yaw error for an idling rotor in high winds, resulting in large edgewise motion and associated large motion in the tower side-side and platform-sway directions. Can you confirm the inflow-skew angle and large edgewise blade deflection?

Best regards,

[Hong.Wu]

Dear @Jason.Jonkman ,

Thank you for your reply. I agree with your assessment that this issue is related to the tower. When I disabled the tower’s side-side and fore-aft DOF, treating it as a rigid body, the wave radiation force stabilized.

Regarding the potential for classic blade-edgewise instability from an idling rotor in high winds, I have re-run the simulation in OpenFAST. The yaw degree of freedom has been disabled in all my simulations.

Here are the results for the tower side-side motion, platform-yaw and sway motion.

the results of flapwise, edgewise blade tip deflection and inflow-skew angle as follows,

I think that while the sway and tower top side-side motion itself is not large in amplitude, but it is highly fluctuation (at 3.7 Hz), which is agree with the radiation force.

I’m seeing that the wave radiation force starts to become unstable around 800 seconds, and this happens before the lager blade tip edgewise deflection. So I am uncertain whether this is a case of blade edgewise instability or rather a tower resonance. Inspired by your thesis (NREL 5MW), it seems to me that the 3.7 Hz frequency is more closely aligned with the tower’s second natural frequency. However, I’m still unclear on how this instability is occurring, as my first-order wave forces do not contain this high-frequency component.

I look forward to your suggestions.

Best regards,

[Jason.Jonkman]

Dear @Hong.Wu,

I’m not seeing very large rotor skew, but it could be that the blade-edgewise instability is starting to happen at the lower skew of your simulation.

My understanding is that you are simulating the IEA Wind 15-MW RWT atop the VolturnUS-S semisubmersible in an idling condition, so, I’m not sure why you are sharing natural frequencies of the land-based NREL 5-MW turbine. I’m not familiar with the full-system natural frequencies of the IEA Wind 15-MW RWT atop the semisubmersible, but I would suggest figuring these out so that you can properly identify the natural frequencies in the PSDs.

I presume you are using BeamDyn to model the blade structural dynamics rather than ElastoDyn. My guess is the 3.7 Hs is the 3rd edgewise bending mode or 2nd tower bending mode, but you could perform an OpenFAST linearization analysis using ACDC to confirm.

Best regards,

[Hong.Wu]

Dear @Jason.Jonkman ,

Yes, my simulation involves the IEA 15MW RWT (Reference Wind Turbine) atop a floating foundation in an idling condition. My hypothesis is that the 3.7 Hz frequency I’m observing corresponds to the 2nd tower side-to-side bending mode. I plan to verify this by performing a linearization analysis using OpenFAST. Thank you for your patient guidance and insightful answers.

Best regards,