BModes : Input parameters about tower support subsystem

Wind & Water Computer-Aided Engineering Software Tools
41

Dear Jason,

I am currently looking at the IEA 15 MW turbine, I am calculating the natural frequency using BModes. Could I know how can I find the cm_loc, cm_axial, ixx_tip, iyy_tip, izz_tip, ixy_tip, izx_tip, iyz_tip" which are input for the .bmi file.

Thanks,
Satish J

42

You can find those quantities in Table 5-1 in the IEA Wind 15-MW report. However, there is a known issue with the summation of the moments of inertia (see Github Issue here). I am working on a more reliable estimate, but it will be a bit before it is ready to go.

43

Dear Jason,

I am trying to better understand the body-weight restoring into elements (4,4) and (5,5) of hydro_k in Bmodes.

I see from WAMIT’s manual that C(4,4) and C(5, 5) contain -mgz_g where m is supposedly the platform mass, and z_g its CM z coordinate.
So if we modify these terms by adding -MtotgZ_Gtot, (Mtot=mass of the entire system except for mooring), wouldn’t it double count the contribution from the platform?
Thank you in advance.

44

Dear Rick,

You should use the full-system mass and center of mass to derive the hydro_k in BModes (what you refer to as Mtot and Z_Gtot). The WAMIT documentation implies this as well, as it is generally referring to oil and gas platforms, whose platform mass is effectively the full-system mass.

Best regards,

45

Dear Jason,

I have modelled both tower and flexible foundation of an Onshore WT in SubDyn. And I got relevent mass and stiffness matrices. Now I am trying to find tower Modes using BModes.

Tower height= 136.8 m
SubDyn’s Total Mass (structural and non-structural)= 0.121917E+07
SubDyn’s Total Mass CM coordinates (Xcm,Ycm,Zcm) = (0, 0, 40.5)

What should be my draft, cm_pform, mass_pform, ref_msl in BModes?

I have given some lumped masses when I modelled the Tower in SubDyn. Also should I give the whole tower again (along with lumped masses at different sections) in Tower section properties when running BModes?

Thank you in advance.

46

Dear @Tom.Jose,

I’m not sure I fully understand your question. BModes is most useful when calculating tower mode shapes for ElastoDyn. But it sounds like you are modeling the entire support structure (tower + foundation) in SubDyn, so, presumably the tower degrees of freedom of ElastoDyn are disabled. What are you trying to use BModes for?

Best regards,

47

Dear Jason,

I am currently working to reproduce the tower mode shape polynomial fit coefficients used in the OpenFAST OC3Hywind project.

During my analysis, I found discrepancies in the 2nd order tower mode shape between what I obtained using BModes_JJ and the reference values provided in OpenFAST. Additionally, the polynomial fit coefficients calculated from BModes mode shapes also differ from those in the OpenFAST input file, as well as those derived from the ModeShapeFitting spreadsheet.

Below is a comparison of the mode shapes obtained from the OpenFAST case file, BModes simulation, and ModeShapeFitting.xls (where BModes output was used as input).

Mode Shape Comparsion
Mode Shape Comparsion957×1205 58.6 KB

（All mode shapes shown are normalized by the tower top value, and a zero-base offset has been applied.）

For tower mode shape analysis, I used the “OC3Hywind.bmi” file inside the package of BModes_JJ you shared.

As you mentioned earlier, I interpreted the 7th and 8th BModes output as the 1st order tower mode, and the 10th and 11th outputs as the 2nd order tower mode, as shown below:

  eigenvalue(  7) =  0.834905D+11        mode  7 frequency =      0.481579
  eigenvalue(  8) =  0.867022D+11        mode  8 frequency =      0.490754

  eigenvalue( 10) =  0.151650D+13        mode 10 frequency =      2.052434
  eigenvalue( 11) =  0.234819D+13        mode 11 frequency =      2.553970

The reference polynomial coefficients for the tower are from

5MW_OC3Spar_DLL_WTurb_WavesIrr\NRELOffshrBsline5MW_OC3Hywind_ElastoDyn_Tower.dat

I am wondering if there might be any mistake in my process, or if these differences are expected due to variations between the “OC3Hywind.bmi” and the “NRELOffshrBsline5MW_OC3Hywind_ElastoDyn_Tower.dat” configurations, such as differences in boundary conditions, normalization method, or model parameters.

Any insights or suggestions would be greatly appreciated.

Thank you very much in advance!

Best regards,
Yunpeng Song

48

Dear @Yunpeng.Song,

This question has been asked and answered on this forum before–e.g. see the posts dated May 07-08, 2018 in the following forum topic: Ansys model tower - #12 by Jason.Jonkman.

Best regards,

49

Dear Jason,

Thank you very much for your quick response.
I understand the reason now.

Best regards,
Song