How to apply icedyn to floating turbines in FAST

Wind & Water Computer-Aided Engineering Software Tools
1

Hi all,

I am currently using glue-code in openfast for debugging calculations. Take 5MW_OC3Spar as an example, I can call icedyn in the setup file, but I need subdyn which is not contained in the 5MW_OC3Spar folder . How do I get the subdyn file, do I need to design the modeling for this?

Kind regards,
Liu.

2

Dear Liu,

Until a recent pull request into the dev branch of OpenFAST, SubDyn was only applicable to fixed-bottom, not floating systems. However, the following pull request allows SubDyn to be enabled for floating offshore wind turbines, which is applicable to modeling floating substructure flexibility and the calculation of internal member-level loads: github.com/OpenFAST/r-test/issues/39. (The testing we have done has been for floaters that contain proprietary information that cannot be shared publicly.) If you wish to introduce a SubDyn model of the OC3-Hywind spar, you’ll have to switch to the dev branch of OpenFAST and you’ll likely have to make your own assumptions about the structural characteristics of the spar (as these details were never published).

Best regards,

3

Dear Jason

What is the difference between the ‘RotSpeed’ and the ‘LSSGagVxa’ ?

4

Dear Rui.Hao,

ElastoDyn output RotSpeed is equivalent to the ElastoDyn output LSSTipVxa (LSSTipVxa and RotSpeed are two different names for the same output), which is the rotational speed of the low-speed shaft on the rotor side of the shaft. LSSGagVxa is similar, but LSSGagVxa is the rotation speed of the low-speed shaft on the gearbox side of the shaft. The difference between these two is the flexibility of the drivetrain, if enabled (DrTrDOF = TRUE). If the drivetrain flexibility is disabled (DrTrDOF = FALSE), then LSSGagVxa = LSSTipVxa = RotSpeed.

Best regards,

1 Like
5

Dear @Jason.Jonkman

Whether do I enable DrTrDOF or not, there is difference between LSSGagVxa and HSShftV. Could you please explain what is the cause? Is that due to the existence of the gearbox? Furthermore, from the experiment of NREL 5MW in OpenFAST, I found that the difference between LSSGagVxa and HSShftV is much larger than the difference between LSSGagVxa and RotSpeed, could you please explain that?

What is more, do we consider all the torsional factors(spring torque and damping torque)to low-speed shaft in OpenFAST? To be more specific, the high-speed shaft is rigid in OpenFAST?

Thanks in advance!

6

Dear @Yinghan.Liu,

Just a few comments:

  • The drivetrain in ElastoDyn is modeled with rigid-body rotation and a torsional spring, expressed for the low-speed shaft. This doesn’t mean that all torsional flexibility has to be within the low-speed shaft (torsional flexibility of the shaft could come from a combination of the low-speed shaft, high-speed shaft, and gearbox), just that the model and parameters must be specified relative to the low-speed shaft.
  • The virtual strain gauge on the low-speed shaft is assumed to be located on the generator-side of the spring, with the hub/rotor on the other side of the spring.
  • The high-speed shaft speed is modeled simply by multiplying the low-speed shaft speed (on the generator-side of the spring) by the gearbox ratio (GBRatio).
  • Thus, LSSGagVxa = GBRatio * HSShftV and LSSGagVxa differs from RotSpeed by the small contribution from the rotation of the spring.
  • The following forum topic is likely of interest to you: Resistant moment of the rotor and of the electric generator.

Best regards,

7

Dear @Jason.Jonkman

Thanks so much for your quick reply!

There is one thing I found strange. In OpenFAST, from the results I found that HSShftV is not equal to GearboxRatio * LSSGagVxa, why that happens?

Moreover, do we ignore the inertia of the gearbox? Have we count the stiffness of the gearbox into the low-speed shaft or not?

Looking forward to your reply.

8

Dear @Yinghan.Liu,

I would expect HSShftV to equal LSSGagVxa * GBRatio; please clarify what you are seeing otherwise.

The full torsional flexibility of the shafts and gearbox should be lumped together and expressed in DTTorSpr (expressed relative to the low-speed shaft) and the full rotational inertia of the shafts, generator, and gearbox should be lumped together and expressed in GenIner (expressed relative to the high-speed shaft) and HubIner (expressed relative to the low-speed shaft).

Best regards,