5 MW barge type FOWT (ITI Energy Barge) natural frequencies

Wind & Water Computer-Aided Engineering Software Tools
1

Hello everyone,

I am working now on the 5 MW barge type FOWT (ITI Energy Barge). I was running decay tests in order to verify the natural frequencies presented in section 5.1 in Matha PhD Thesis (“Model Development and Loads Analysis of an Offshore Wind Turbine on a Tension Leg Platform, with a Comparison to Other Floating Turbine Concepts”).

The decay tests were ran using OpenFAST v3.1
In .fst file, i disabled AeroDyn, Inflow, ServoDyn. In Hydrodyn, i set WaveMod = 0 in order to set “still water” condition. In ElastoDyn file, all DoF are enabled except for NacelleYaw and Teeter.

I obtained important discrepancies in surge, sway and tower 1st mode in side-side. In addition for the heave motion, i used to compute the PSD in log scale to see a little peak at the natural frequency 0.136 Hz. In the linear scale, this peak does not appear. I used MATLAB to compute PSD. For the surge, sway and heave, an initial displacement equals 1 m is used and for the roll, pitch and yaw an initial angle of 1 deg is used. Those values are chosen to remain around the static equilibrium position where the nonlinearities of the mooring lines are considered negligible. Also, for the surge, sway and heave, i ran a decay test for 8000 s and 1500 s for the others.

In the following, a picture shows the natural frequencies obtained versus those present in Matha et al., 2009.

Capture
Capture1107×128 3.92 KB

Does anyone encounter the same difference or at least tell me where i did the mistake.

Also, i wanna make sure that 3P (self excitation) excites for example the tower in fore-aft only ? or also in side-side ?

Thank you for your help.

Best Regards,

Riad

2

Dear @Riad.Elhamoud,

I’m not sure I know where these discrepancies arise from (I have not used this model in any great depth in many years). Just a couple comments:

  • For the tower side-side mode, the generator degree of freedom plays a role (because it will impact the tower-top inertia). Does your result improve relative to Matha et al if you disable generator DOF (GenDOF = FALSE)?
  • For the surge and sway modes, the stiffness is mainly from the mooring systems. Is the natural frequency sensitive to the displacement?

Best regards,

1 Like
3

Dear @Jason.Jonkman,

Thank you for your reply.

:black_medium_small_square: I have tried to disable the generator DoF and it works !!! The result now is improved relative to Matha et al. In the following, the PSD of the ttdspss is presented. In the PSD, one could notice two peaks which correspond to platform roll frequency and tower 1st bending mode in side-side direction. This is reasonable !

twrss_decay_test_gen_dof_disabled
twrss_decay_test_gen_dof_disabled700×525 15.5 KB

However, in reality, the generator DoF should not be disabled.

:black_medium_small_square: I did not make a sensitivity analysis, but mooring lines are not linear. Indeed, their stiffness changes with the displacement and this should affect the natural frequencies of the platform. There will be a little change in the values of the frequencies. Like, when the mooring lines are replaced by mooring matrix which are valid only in the vicinity of the equilibrium point. Do you agree ?

Best Regards,

Riad

4

Dear @Riad.Elhamoud,

A stiffness matrix is a linearized approximation of a mooring system, representing the effect of small displacements about an operating point.

Best regards,

1 Like