SubDyn simulation Node force output problems

[HiHabi]

Dear all,
I'm currently working on the Openfast simulation and the comparison between the abaqus model. As I'd modify parts of the Jacket Structure in SubDyn and the entire simulation is idea, however when outputting the dynamic force at certain member as well the nodes, it always displayed 0 for every output node I'd set in the SubDyn inputfile, can anyone help with this problem?

the main task of my problem is to print the dynamic force at the Xjoint, where the nodeID=28, 30
the connecting memberID sets are as follow
MemberID/ Node1ID/ NodeID2/ MPropSetID1/ MPropSetID2/ MType
31 8 28 5 5 1
32 14 28 5 5 1
39 9 28 6 6 1
40 15 28 6 6 1
63 10 30 7 7 1
64 16 30 7 7 1
71 11 30 7 7 1
72 17 30 7 7 1

------------------------- MEMBER OUTPUT LIST ------------------------------------------
8 NMOutputs - Number of members whose forces/displacements/velocities/accelerations will be output (-) [Must be <= 9].
MemberID NOutCnt NodeCnt ![NOutCnt=how many nodes to get output for [< 10]; NodeCnt are local ordinal numbers from the start of the "member," and must be >=1 and <= NDiv+1] If NMOutputs=0 leave blank as well.
(-) (-) (-)
31 1 3 ! M1
32 1 3 ! M2
39 1 3 ! M3
40 1 3 ! M4
63 1 1 ! M5
64 1 1 ! M6
71 1 5 ! M7
72 1 5 ! M8
------------------------- SDOutList: The next line(s) contains a list of output parameters that will be output in .SD.out or .out. ------
"IntfFXss, IntfFYss, IntfFZss, IntfMXss, IntfMYss, IntfMZss" - Total interface reaction forces and moments at the TP reference point location in SS coordinate system
"M5N1FKXe, M5N1FKYe, M5N1FKZe, M5N1MKXe, M5N1MKYe, M5N1MKZe"
"M6N1FKXe, M6N1FKYe, M6N1FKZe, M6N1MKXe, M6N1MKYe, M6N1MKZe"
"M7N5FKXe, M7N5FKYe, M7N5FKZe, M7N5MKXe, M7N5MKYe, M7N5MKZe"
"M8N5FKXe, M8N5FKYe, M8N5FKZe, M8N5MKXe, M8N5MKYe, M8N5MKZe"
END of output channels and end of file. (the word END must appear in the first 3 columns of this line)

The outputs for all M7N5 and M8N5 are all zeros as second figure.

Please offer any suggestion about fixing this problem, many thanks.

Sincerly,
Jack

the following is the modified SubDyn file, the Openfast simualtion seem terminated normally yet the output force from the node are always remain 0.
SybDyb_backup.txt

[jjonkman]

Dear @HiHabi,

I see two problems with your SubDyn input file. First, you've only selected one output node (i.e., NOutCnt = 1) on each output member, so, you can only select N1 in the SDOutList (N2-N9 are invalid and will return zeros). Second, your model uses only two elements per member (NDiv = 2), so each member will have 3 nodes. So, you can't set NodeCnt higher than 3 (whereas you have set some of them to 5).

Best regards,

[HiHabi]

Dear @jjonkman

Thanks for your kind response,
I'd see where the mistakes I'd made, thanks for the clarification.

As I'm understood, as taking the MemberID=71 for instance, where the NodeID2 of the member is 30 as I'd like to output the force values, the modification of the parameters are:
NDiv=5 (the assuming subelements I'd like to simulate, does this makes huge difference between 2 or 5?)
NodeCnt=6, refering from the NDiv=5+1 to print the response at NodeID=30 as this is the final node in the member?

please help with the confusion, many thanks

Sincerly, Habi

[HiHabi]

Dear @jjonkman

As I'd remain the NDiv=2 for testing, the following pic are the output list of the member force being modified. The adjusted SubDyn input file as well the openfast simulation error are as follow:

SybDyb_backup.txt

Still taking the MemberID=71, with NodeID1=11/ NodeID2=30, if I'd like to output the dynamic force at NodeID=30, would you recommend to swap the NodeID in MemberID=71 as NodeID1=30/ NodeID2=11 as solution? Or how may I adjust the output list properties with the valus setting for parameters as NOutCnt/ NodeCnt?

Many thanks for your precious time,

Sincerly, Jack

[jjonkman]

Dear @HiHabi,

NodeCnt is simply the a list of nodes from 1 to NDiv+1. When NDiv=2, the member has two elements with three nodes, so, NodeCnt can be set to 1 (start node), 2 (middle node), and/or 3 (end node). For member 71, the start node is 11 and end node is 30, so, NodeCnt = 1 corresponds to NodeID=11 and NodeCnt=3 corresponds to NodeID=30.

Best regards,

[HiHabi]

Dear @jjonkman

It seem my understandings are correct, however is there a must of changing at the parameter NOutCut from 1 to 3?
I'd test on this parameter for NOutCut=1/ 2/ 3 while remaining the NodeCnt=3 for Member7, as the pattern as M7N3FKXe and so, yet the error of Failed to read member output list properties still shows up in NOutCut=2 and 3, only when the NOutCut=1 that Openfast can simulate properly yet this turn back to the output zero problems as the begin.

Please help with this problem, many thanks.

[jjonkman]

Dear @HiHabi,

NOutCnt is the number of output nodes for the specific output member. If NOutCnt > 1, you must identify that many values of NodeCnt. For example, if you set NOutCnt=2 with NodeCnt=1,3 for MemberID=71, i.e.: 71 2 1,3, then output node 1 will correspond to NodeID=11 of member 71 and output node 2 will correspond to NodeID=30 of member 71.

Best regards,

[HiHabi]

Dear @jjonkman ,

Thanks again for your kind reply and patients.

As I'd follow your instruction about adjusting the memberoutputlist and the output results are still remaining as zero as I'd like to output the response at NodeID=30 at memberID=71 and 72, as M7N3 and M8N3 from the memberoutput list.

The NodeID=30 is setted as the 3rd layer Xjoint for the jacket substruct for your reference, however I'd believe the response of this place would not be zero.

however, if the simulation is submitted with M7N1 and M8N1, it seems work properly as the output of the node dynamic force aren't zero, therefore I believe the problem doesn't result from structure modeling problem.
Would you please suggest any idea about solving this problem?

Results for M7N1 and M8N1:

Results for M7N3 and M8N3:

[jjonkman]

Dear @HiHabi,

You've defined NOutCnt = 2, so, there are only 2 output nodes per member. Thus, you can select N1 and N2 in the member output names within the SSOutList, but N3 is invalid and will return zeros. N2 refers to the 3rd node along the member per your setting.

Best regards,

[HiHabi]

Dear @jjonkman

Thanks for your kind reply, the output results are available now.
However I still have some confusion as follow:

For the Xjoint Structure, i.e. NodeID=30, the surroudning MemberID=63/ 64/ 71/ 72

MemberID MJointID1 MJointID2 MPropSetID1 MPropSetID2 MType
63 10 30 7 7 1
64 16 30 7 7 1
71 11 30 7 7 1
72 17 30 7 7 1

As I print output all the Dynamic response of all members mentioned above as the output node format as:
MemberID/ NOutCnt/ NodeCnt
63/ 2/ 1,3
64/ 2/ 1,3
71/ 2/ 1,3
72/ 2/ 1,3

the responses for each member at NodeID=30 seem varies a lots, is this the correct result for the structure response? As my assumptions are that the surround members output at NodeID=30 shall share similar response for convergence.
For example, F5N2FMXe shall share similar results with F6N2FMXE, F7N2FMXe and F8N2FMXe as all these points are referring to the FX dynamic result at NodeID=30.
The Dynamic responses are give as follow pic, as well the excel file, second sheet.

Node30 4member response.xlsx

Please help with this problem, many thanks.

[jjonkman]

Dear @HiHabi,

I haven't looked closely at your results, but these loads are measured in different members with different local coordinate systems, so, I would not expect the loads to be the same.

Best regards,

[HiHabi]

Dear @jjonkman

The reason that I'd like to measure the node dynamic force is that I'd create another identical model in abaqus, as I'd like to make some work on comfirming the two results having similar result and making some comparison, that is the main object I'd like to work on the output of NodeID=30.

Is there any suggestion you'd suggest to make the verification, or should I try to make the coordinate transformation for each member results above into global coordinate ones' for comparison.

Sincerely, Jack

[jjonkman]

Dear @HiHabi,

The coordinate system is only one reason for the difference. The other reason is that different members will have different loads, even when the intersect at a joint.

Can't you compare the member-level loads directly between SubDyn and ABAQUS for verification?

Best regards,

[HiHabi]

Dear @jjonkman

As I'd closed out the hydrodyn for the aero elastic oneway coupled simulation in Openfast as well in Abaqus, the displacement for observation points seem matching, however with the hydrodyn added, the difference may vary to a large range. Therefore a further quesiton about how to apply the response of hydrodyn into abaqus will be the further questions I'd like to seek for your advices, is there any suggestiong that I may as well applied the hydrodyn results into the abaqus software that can sharing the similar dynamic behaviours as hydrodyn to Openfast main software?

Thanks for your kind reply and patience,
Sincerely, Jack

[jjonkman]

Dear @HiHabi,

In OpenFAST, the hydrodynamic loads are dependent on the structural motion (hydroealastic coupling). So, it would not be appropriate to take the hydrodynamic loads from OpenFAST and apply them directly to ABAQUS. Rather, a two-way hydro-elastic coupling would be required.

Best regards,