OC4 Phase II Load case 3.9 Flooded Column

Wind & Water Computer-Aided Engineering Software Tools
1

Hello everyone,

I am trying to reproduce the load case 3.9 of a flooded column from the OC4 Phase II study (document attached).

In part 2.3 Full-System Analysis, page 4, there is the description of the 2 methods to model this damage case:

  1. directly modelling the additional volume of water in the base column 1 and upper column 1,
  2. or changing the platform properties to represent the additional weight.

For the first method, I understand I have to modify the Filled Members part of the HydroDyn input file, based of the geometrical information of Figure 1, page 5 of the attached document. Though, in the Filled Members part, the Filled Free Surface Location (FilledFSLoc) is written in z-axis coordinates, unlike the geometrical properties in Figure 1. So I guess I need to find somewhere the lowest coordinates of the base column 1 and of the upper column 1 to respectively write in FilledFSLoc:

  • lowest z coordinate of the base column 1 + base column height (6 m) - 2*thickness of the base column 1
  • lowest z coordinate of the base column 1 + base column height (6 m) + 9.33 m

Is this calculation right? Is the lowest z coordinate of the base column 1 in the Jointzi of the corresponding member? If not, where can I read the lowest z coordinate of the base column 1 please?

As for the second method, changing the properties of the platform, I don’t know where to modify the Ixy, Izx, and Iyz parameters. Also, it is said “relative to old CM”, so it means I have to add the values figured in Table 5 for this document to the current values of my input files, or just replace it by those values?

Thank you for your help,

Have a nice day/weekend,
Bertrand
NREL - OC4 Phase II DeepCWind - Load Cases.pdf (583 KB)

2

So for the first method, using the geometrical data from the z coordinates of the lowest joints of the base column and upper column members, I have found the following results for the Filled Members part (member 2 is the upper column 1, member 5 is the base column 1):

---------------------- FILLED MEMBERS ------------------------------------------
4 NFillGroups - Number of filled member groups (-) [If FillDens = DEFAULT, then FillDens = WtrDens; FillFSLoc is related to MSL2SWL]
FillNumM FillMList FillFSLoc FillDens
(-) (-) (m) (kg/m^3)
2 3 4 -6.17 1025
2 6 7 -14.89 1025
1 2 -4.67 1025
1 5 -13.94 1025

…with the following Members data (just including the base and upper columns, did not copy all of the member’s list):

-------------------- MEMBERS -------------------------------------------------
25 NMembers - Number of members (-)
MemberID MJointID1 MJointID2 MPropSetID1 MPropSetID2 MDivSize MCoefMod PropPot [MCoefMod=1: use simple coeff table, 2: use depth-based coeff table, 3: use member-based coeff table] [ PropPot/=0 if member is modeled with potential-flow theory]
(-) (-) (-) (-) (-) (m) (switch) (flag)
1 1 2 1 1 1.0000 3 TRUE ! Main Column
2 3 4 2 2 1.0000 3 TRUE ! Upper Column 1
3 5 6 2 2 1.0000 3 TRUE ! Upper Column 2
4 7 8 2 2 1.0000 3 TRUE ! Upper Column 3
5 42 3 3 3 1.0000 3 TRUE ! Base Column 1
6 43 5 3 3 1.0000 3 TRUE ! Base Column 2
7 44 7 3 3 1.0000 3 TRUE ! Base Column 3

…so JointID1 of the base column is number 42, and 3 for the upper column,

…and with the Member Joints list:

---------------------- MEMBER JOINTS -------------------------------------------
44 NJoints - Number of joints (-) [must be exactly 0 or at least 2]
JointID Jointxi Jointyi Jointzi JointAxID JointOvrlp [JointOvrlp= 0: do nothing at joint, 1: eliminate overlaps by calculating super member]
(-) (m) (m) (m) (-) (switch)
1 0.00000 0.00000 -20.00000 1 0
2 0.00000 0.00000 10.00000 1 0
3 14.43376 25.00000 -14.00000 1 0
4 14.43376 25.00000 12.00000 1 0
5 -28.86751 0.00000 -14.00000 1 0
6 -28.86751 0.00000 12.00000 1 0
7 14.43376 -25.00000 -14.00000 1 0
8 14.43376 -25.00000 12.00000 1 0
9 14.43375 25.00000 -20.00000 2 0
10 -28.86750 0.00000 -20.00000 2 0
11 14.43375 -25.00000 -20.00000 2 0
12 9.23760 22.00000 10.00000 1 0
13 -23.67130 3.00000 10.00000 1 0
14 -23.67130 -3.00000 10.00000 1 0
15 9.23760 -22.00000 10.00000 1 0
16 14.43375 -19.00000 10.00000 1 0
17 14.43375 19.00000 10.00000 1 0
18 4.04145 19.00000 -17.00000 1 0
19 -18.47520 6.00000 -17.00000 1 0
20 -18.47520 -6.00000 -17.00000 1 0
21 4.04145 -19.00000 -17.00000 1 0
22 14.43375 -13.00000 -17.00000 1 0
23 14.43375 13.00000 -17.00000 1 0
24 1.62500 2.81500 10.00000 1 0
25 11.43376 19.80385 10.00000 1 0
26 -3.25000 0.00000 10.00000 1 0
27 -22.87000 0.00000 10.00000 1 0
28 1.62500 -2.81500 10.00000 1 0
29 11.43376 -19.80385 10.00000 1 0
30 1.62500 2.81500 -17.00000 1 0
31 8.43376 14.60770 -17.00000 1 0
32 -3.25000 0.00000 -17.00000 1 0
33 -16.87000 0.00000 -17.00000 1 0
34 1.62500 -2.81500 -17.00000 1 0
35 8.43376 -14.60770 -17.00000 1 0
36 1.62500 2.81500 -16.20000 1 0
37 11.43376 19.80385 9.13000 1 0
38 -3.25000 0.00000 -16.20000 1 0
39 -22.87000 0.00000 9.13000 1 0
40 1.62500 -2.81500 -16.20000 1 0
41 11.43376 -19.80385 9.13000 1 0
42 14.43376 25.00000 -19.94000 1 0
43 -28.86751 0.00000 -19.94000 1 0
44 14.43376 -25.00000 -19.94000 1 0

…and also according to the thickness of the base column in the Member Cross-Section Properties:

---------------------- MEMBER CROSS-SECTION PROPERTIES -------------------------
4 NPropSets - Number of member property sets (-)
PropSetID PropD PropThck
(-) (m) (m)
1 6.50000 0.03000 ! Main Column
2 12.00000 0.06000 ! Upper Columns
3 24.00000 0.06000 ! Base Columns
4 1.60000 0.01750 ! Pontoons

…and with the 9.33 m of filled water for the upper column 1 in Figure 1 of the document attached above.

Does the filled members I have modified seem right now? If yes, good news. For curiosity, I still would like an answer for the second method question I asked above.

Thank you for your help,
Bertrand

3

Dear Bertrand,

I agree with your settings to flood the upper column.

For the flooding of the base column, I would think you’d want to use FillFSLoc = -14.06 m (instead of -13.94 m or -14.89 m), which is the depth of the base column top cap.

For method 2, the center of mass stated in Table 5 in your attachment is relative to the original CM, which was (0,0,-13.46) m, so the new CM is (0.5684, 0.9847, -13.38998) m. The mass and inertia values in Table 5 in your attachment are absolute. These mass, center of mass, and inertia values can be specified in the ElastoDyn primary input file (PtfmCMxt, PtfmCMyt, PtfmCMzt, PtfmMass, PtfmRIner, PtfmPIner, PtfmYIner), but ElastoDyn does not consider the cross inertia terms. Of course, if you set these, you’d have to remove the water ballast from HydroDyn altogether.

Best regards,

4

Dear Jason,

Thank you for your clear answer.

Indeed, I made a mistake for the flooding of the base column 1, it should be FillFSLoc = -14.06 m. Though, only column 1 is flooded, so I would let FillFSLoc = -14.89 m for base columns 2 and 3.
By the way, do you know why upper column 1 is flooded by this specific 9.33 m? Why not less, or more, or even all the height of the upper column?

Okay, I got it for method 2. Although, the cross inertia terms should be modified, but we cannot? So this method is less reliable maybe?
Also, I had different default coordinates for the CM, still for the DeepCWind platform:
0 PtfmCMxt - Downwind distance from the ground level [onshore] or MSL [offshore] to the platform CM (meters)
0 PtfmCMyt - Lateral distance from the ground level [onshore] or MSL [offshore] to the platform CM (meters)
-8.6588 PtfmCMzt - Vertical distance from the ground level [onshore] or MSL [offshore] to the platform CM (meters)
I hope it’s right.

Thanks again for the help,

Best regards,
Bertrand

5

Dear Bertrand,

No, I don’t recall the source for the 9.33 m.

ElastoDyn does not by default include the cross inertia terms, but the source code could be modified to include them. When NREL modeled this load case, we used the water ballasting feature of HydroDyn rather than lumping the ballast into the structural mass/inertia properties in ElastoDyn.

The value of PtfmCMzt = -8.6588 m is for the steel parts of the OC4 semisubmersible substructure, i.e., assuming that the water ballast is accounted for in HydroDyn. If the water ballast is lumped into the structural mass/inertia properties in ElastoDyn, the value reduces to PtfmCMzt = -13.46 m (the platform mass and inertia in ElastoDyn would also change).

Best regards,

6

Dear Jason,

Thank you for your answer, I got it now. :slight_smile:

Best regards,
Bertrand