Simulation error under extreme wave and wind conditions.

Wind & Water Structural Analysis
1

Hi all,

I got some errors when running the simulation using test 24 under the extreme condition as followed:

35.7m/s turbulent wind, Hs = 14.1m, Tp = 14.3s, turbine parked (rotor speed = 0).

According to Dr Jonkman’s article “Dynamics of offshore floating wind turbines—analysis of three concepts”, FAST can run normally under even harsher environment conditions, therefore I guess some modifications might be made somewhere when the loads are extremely high.

Here are some of the error messages I got:
FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption1:ED_HD_InputOutputSolve:HydroDyn_CalcOutp
ut: Angles in GetSmllRotAngs() are larger than 0.4 radians.
ED_HD_InputOutputSolve:HydroDyn_CalcOutput: Angles in GetSmllRotAngs() are larger than 0.4
radians.

FAST_Solution:FAST_AdvanceStates:AD_UpdateStates:BEMT_UpdateStates(node 5, blade
1):BEMT_UnCoupledSolve:There is no valid value of phi for these operating conditions: Vx =
-4.8869, Vy = -240.97, rlocal = 11.757, theta = 0.23226, geometric phi = -3.1213
BEMT_UpdateStates(node 6, blade 1):BEMT_UnCoupledSolve:There is no valid value of phi for these
operating conditions: Vx = -21.538, Vy = -283.77, rlocal = 15.859, theta = 0.20035, geometric phi
= -3.0658
BEMT_UpdateStates(node 7, blade 1):BEMT_UnCoupledSolve:There is no valid value of phi for these
operating conditions: Vx = -40.407, Vy = -329.69, rlocal = 19.956, theta = 0.17734, geometric phi
= -3.0196
BEMT_UpdateStates(node 8, blade 1):BEMT_UnCoupledSolve:There is no valid value of phi for these
operating conditions: Vx = -54.007, Vy = -378.24, rlocal = 24.054, theta = 0.15725, geometric phi
= -2.9998
FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption1:ED_HD_InputOutputSolve:HydroDyn_CalcOutp
ut: Angles in GetSmllRotAngs() are larger than 0.4 radians.
E

FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption1:ED_HD_InputOutputSolve:HydroDyn_CalcOutp
ut: Angles in GetSmllRotAngs() are larger than 0.4 radians.
ED_HD_InputOutputSolve:HydroDyn_CalcOutput: Angles in GetSmllRotAngs() are larger than 0.4
radians.

FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:InflowWind_CalcOutput:CalcOutput:IfW_TSF
FWind_CalcOutput [position=(5986.1, -6780.8, -9175.3) in wind-file coordinates]: Error: FF wind
array was exhausted at 5.15 seconds (trying to access data at -101.59 seconds).
CalcOutputs_And_SolveForInputs:SolveOption1:ED_HD_InputOutputSolve:HydroDyn_CalcOutput: Angles in
GetSmllRotAngs() are larger than 0.4 radians.
ED_HD_InputOutputSolve:HydroDyn_CalcOutput: Angles in GetSmllRotAngs() are larger than 0.4
radians.

FAST encountered an error at simulation time 5.1375 of 3000 seconds.
Simulation error level: FATAL ERROR

Aborting FAST.

I would be very appreciate if anyone could help with this situation.

2

Dear Yajun,

Yes, FAST can be used to simulate this case. Based on the extreme position of an aerodynamic analysis node (5986.1, -6780.8, -9175.3), clearly your model has gone unstable for some reason. You haven’t stated much about how you changed Test 24, but one thing I see that does not look correct is that you’ve enabled blade-element/momentum theory (BEMT) in AeroDyn, but you’ve said the rotor speed is zero. BEMT is only valid for an operational rotor. According to the draft AeroDyn User’s Guide and Theory Manual (openfast.readthedocs.io/en/mast … index.html), for an idling or parked turbine, the AeroDyn settings should: WakeMod = 0 (to disable BEMT) and AFAeroMod = 1 (to disable unsteady airfoil aerodynamics). I would suggest starting by making these changes and rerunning the case.

Best regards,

3

Dear Jason,

Thank you for your timely reply.

For Test 24, I firstly change the Hs and Tp in HydroDyn to 14.1m and 14.3s, then I regenerate a wind file using TurbSim (URef = 35.7m/s, RefHt = 10m). The total simulation period is expended to 3000s. The RotSpeed, GenDOF, PCMod were all changed to 0. I changed BlPitch to 90 degree. According to your suggestions, I set wAKEmOD = 0, AFAeroMod = 1, and got the error messges:

FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:InflowWind_CalcOutput:CalcOutput:IfW_TSF
FWind_CalcOutput [position=(5.24088E+13, -7.08035E+13, -2.95520E+13) in wind-file coordinates]:
Error: FF wind array was exhausted at 8.225 seconds (trying to access data at -9.45993E+11
seconds).
SolveOption2:AD_CalcOutput:SetInputs:TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower strike.
TwrInfl:getLocalTowerProps:Tower str

FAST encountered an error at simulation time 8.2125 of 3000 seconds.
Simulation error level: FATAL ERROR

Aborting FAST.

Besr regards,
Yajun

4

Dear Yajun,

It still looks like your model is going unstable. I’m not sure why you would need to if you have not changed any geometry, stiffness, or mass properties or time steps, but have a tried reducing the time step (e.g., do you get the same error if you drop the time step by a factor of 10)? If that works, I would guess the model is numerically unstable at the larger time step. If that doesn’t work, I would guess the model is physically unstable for some reason.

Best regards,

5

Dear Jason,

Thank you for your reply.

I tried to reduce the time step to 300s, but got the same error messages. When I reduced the wind speed by a half the system terminated normally.

I have not changed any geometry, stiffness or mass properties, did you modify any of them when doing the extreme loading simulations? It seems that the system was trying to read the wind file in a negative time step which does not exist, could you explain what the reason is?

Best regards,
Yajun

6

Dear Yajun,

By time step, I meant FAST input DT, not the total simulation time TMax.

No, I didn’t change any geometry, stiffness, or mass properties.

The error identifies a position of an aerodynamic analysis at a nonsensical (X,Y,Z) = (5.24088E+13, -7.08035E+13, -2.95520E+13) m, which implies a very large deflection. Wind data files are periodic, so, time effectively extends through +/- infinity.

Because reducing the wind speed seems to help, the problem must be aerodynamic related. Your original wind speed was 35 m/s at 10 m. What is the wind speed at hub height based on the shear you are using? What is the turbulence intensity used?

Best regards,

7

Dear Jason,

I reduced DT to 0.125s and it worked. Thank you very much for your help!

Best regards,
Yajun

8

Dear Yajun,

The original DT of Test24 was 0.0125 s. You are one order of magnitude larger than this. Do you have a typo in your post?

Best regards,