Output of loads at different locations at the substructure in SubDyn

Hello,

I work with the NREL 5 MW reference turbine with the OC3 monopile (FAST v8) and I am interested in the loads at different locations at the substructure. To do this, I used SubDyn to output the loads at the nodes using the member output list or using the option “OutAll” = True. This works quite well so far, but when I had a look at the loads, I noticed the following:

The OC3 monopile has 3 members and 4 Joints:

----------------------------------- MEMBERS --------------------------------------
3 NMembers - Number of frame members
MemberID MJointID1 MJointID2 MPropSetID1 MPropSetID2 COSMID
(-) (-) (-) (-) (-) (-)
1 1 2 1 1
2 2 3 1 1
3 3 4 1 1

---- STRUCTURE JOINTS: joints connect structure members (~Hydrodyn Input File)—
4 NJoints - Number of joints (-)
JointID JointXss JointYss JointZss [Coordinates of Member joints in SS-Coordinate System]
(-) (m) (m) (m)
1 0.00000 0.00000 -20.00010
2 0.00000 0.00000 -10.00000
3 0.00000 0.00000 0.00000
4 0.00000 0.00000 10.00000

Since member 1 is defined by joint 1 and joint 2 and member 2 is defined bei joint 2 and joint 3, I assumed that the loads at the end of member 1 should be the same as at the beginning of member 2. That is correct for the moments. For the forces, however, the time series differ partially. In the case of Fz, for example, when comparing M1J2 and M2J1 (see figure), it seems that the curve is shifted along the z-axis. The time series for Fx and Fy do not always match either (see both figures).


Is my assumption that the forces at the ends of the members should match wrong? Or what could be the reason why they do not match? And is there another way of getting the loads at the substructure at different heights (e.g. -20 m, -10 m, 0 m, 10 m)?

Best regards,
Franziska

Dear @Franziska.Schmidt,

Can you first confirm that you see the same issue if you upgrade from FAST v8 to the current version of OpenFAST (currently v3.5.3). I want to ensure the issue you are reporting has not already been fixed.

Best regards,

Dear @Jason.Jonkman,

as you suggested, I simulated the 5MW system with monopile in OpenFAST 3.5.3. I did not change the files except for the simulation time and the output variables to make sure that I did not add a bug by changing them. In the following two pictures you can see that the same effect occurs. (I used the option: “OutAll” = True to output the member end loads).


Best regards,
Franziska

Dear @Franziska.Schmidt,

I asked internally and realized that the method in which the internal weight is accounted for in SubDyn leads to this discrepancy and is a known issue. This issue has now been reported on the OpenFAST GitHub site: Self-weight in SubDyn · Issue #2325 · OpenFAST/openfast · GitHub. I’m not sure how easy it is to fix this in SubDyn, but hopefully this can be addressed at some point.

Best regards,

Dear @Jason.Jonkman,

thank you for clarifying my question. :slight_smile:

Best regards,
Franziska