Two questions about the modelling of the OC3 Hywind model

Wind & Water Computer-Aided Engineering Software Tools
1

Dear Jason:
The first question is :
I am trying to simulated the six DOFs free decay tests of the OC3 Hywind model using the FAST v7 and FAST v8 respectively. However, I found that the results of dynamic responses of OC3 are different, which makes me confused.
The model configurations in FAST v7 and v8 are the same and listed as follows,

  1. All the DOFs of the OC3 platform have been activated.
  2. The initial surge is 4m while other displacements in other DOFs is 0m or 0Deg.
  3. There is no wind speed, wave height and current.
  4. The turbine control model is disabled.
  5. In FAST v8, the ElastoDyn, AeroDyn v15 without aerodynamic loads on tower, HydroDyn and MAP++ are acitvated.

Because the model configurations in the two versions of FAST are almost the same and the results of them ought to have been consistent. Attached is the comparison of the results (Fig. 1). I don’t know which one is right and how to solve the problem. Hence, I resort to your help. Could you help me? Thanks!

The second question is:

In fact, I am trying to use the CFD code to simulate the dynamic response of the OC3 Hywind model. My modelling steps are,

  1. I caluclate the linearized matrix, such as the M, C, K of the OC3 using FAST V7.
  2. Using the motion equation (see Fig. 2) to calculate the response of the OC3 based on the derived M, C and K, where the aerodynamic loads, hydrodynamic loads are calculated by the CFD code.

I am not sure if it is the right way to develop the CFD model. Could you give me some advice? Thanks!
Fig.2.jpg
Fig. 1.tif (74 KB)

2

Dear Yichau,

It looks like your FAST v8 model has more hydrodynamic damping than your FAST v7 model. I know that the OC3-Hywind model has invoked “additional linear damping”, as documented in the OC3-Hywind specifications report: nrel.gov/docs/fy10osti/47535.pdf. This damping is easily added to the FAST v8 model through the setting of appropriate input parameters in the HydroDyn input file. However, in FAST v7, this requires a slight customization of the source code. Perhaps you have this additional linear damping in FAST v8 and not FAST v7?

Regarding your CFD solution, your equation of motion appears to be missing the hydrodynamic add mass (M_wave) and hydrostatic restoring (K_wave). Regardless, I guess the key here is to not double book terms. If you are modeling the fluids in CFD, you’ll only want to consider the structural terms in the M, C, and K derived from FAST for use in the CFD solution. That is, you’ll want only the structural mass, center of mass, and inertia, and the structural (gravitational) restoring associated with the structural mass.

Best regards,

3

Dear Jason:

Thanks for your help. Following your suggestions, I am considering the structural terms of M, C, K derived from FAST v7. To the best of my knowlege, the linearized matrix K includes the aerodynamic term (K_aero), the hydrostatic term (K_hydrostatic), the gravitational term (K_grav) and the mooring force term (K_mooring). In order to substract the K_aero and K_hydrostatic,

firstly, I disabled the flag of “compute aerodynamic forces” and other flags mentioned in the FAST v7 user guide before do the model linearization.

Secondly, I substracted the K_hydrostatic (given in the “hst” file in the HydroData folder) from the linearized matrix K.

1 ) But I am not sure if it is the right way to obtain the K without K_aero and K_hydrostatic. Could you help me?

  1. In fact, I don’t know how to deal with the linearized matrix M derived from FAST v7, aiming to make sure the M does not contain the hydrodynamic add mass (M_wave). However, according to the linearized stat matrices shown in the FAST user guide, it seems that M is the structral mass matrix without add mass. I am confused and hence resort to your help. Coud you give me some advice? Thank you once again!
4

Dear Yichau,

Here are my answers to your questions:

  1. Yes, by disabling aerodynamics (CompAero = False), you will eliminate K_aero from the FAST v7 linearized solution. And by subtracting off K_hydrostatic (from the WAMIT *.hst file) from the linearized stiffness matrix, you should only have K_grav, and K_mooring remaining (assuming you’ve enabled gravity and moorings and the model is in equilibrium before linearizing).

  2. The mass matrix from the FAST v7 linearized solution will include the effects of hydrodynamic added mass if that is enabled in your model (e.g. from the WAMIT *.1 file). To eliminate this term, use a WAMIT *.1 file with zero added mass.

Best regards,