Forum

Please or Register to create posts and topics.

Modelling a substructure only

Hi David

For my project, I want to model an offshore jacket subjected to wave loads. The turbine on top of it should be represented by a lumped mass at the top of the jacket and no wind loads should be taken into account. Starting from the example files for a jacket, I already constructed the jacket I am investigating with the subelements, but at the moment, there is a wind turbine on top.

I was wondering if it is possible to only model the substructure, so without a tower, blades and so on.
If so, how should I change the .sim and .str files?
I tried a few things, for example putting the number of blades to zero, but then the Python script returns an error (Solver setup failed).

Thanks in advance!
Kind regards
Jelle

Hi Jelle,

it is possible to model only the substructure by setting the substructure file directly as the main structural input file (in this case you dont need the main file at all).

In this way no tower or blades are included in the model and you can set the total lumped inertia of your system via the SUB_MASS table.

You could also set the number of blades to zero in the main structural file as you proposed. The crash that you are seeing is probably caused by the turbine controller. The controller needs to be deactivated if there are no rotor blades, or the .dll will cause a crash).

BR,

David

JelleP has reacted to this post.
JelleP

Hi David

Thank you so much for your help! I was able to model the substructure as I wanted thanks to your explanation.

Now I am working on the post-processing (with the results exported as ASCII), and I would like to relate the local coordinate systems of the subelements to the global coordinate system.
Is there a way to get the local coordinate system orientation in terms of the global coordinate system?

Thanks in advance!
Kind regards
Jelle

Hi Jelle,

great to hear that you have made some progress.

Unfortunately, the element loads are only available in their local coordinate systems. The main purpose being that the bending and shear loads of each element are mainly relevant in the local element frame of reference when it comes to assessing the lifetime of a structural part.

It is also not possible to extract the current orientation of a substructure element from the simulation, however gloabl positions etc. can be accessed.

Technically, it would be very easy to obtain these outputs and add them to the data that is generated, but I don’t see a good reason to do this at the moment, and generally try to keep the amount of data that is being generated relatively small, as this comes at the cost of RAM and HDD memory when projects and result are stored.

Maybe, if you can explain what you would need these outputs for I can be convinced to include those as optional output data?

Cheers,

David

Hi David,

As you also mentioned, the global loads are not really my interest here. It is the deformation of the structure that I am interested in.

I would like to plot the (new) location of every ‘sensor’ to visualize the global deformation. This is easily done with the locations that are saved in the global coordinate system, but since the deformations are very small, I wanted to exaggerate them. Therefore, I wanted to use the local deflections and multiply them by some constant value. However, since these deflections are in the local coordinate system, this does not work.

While I was writing this, I realized I can also use the new global positions relative to the old global positions to get the global deflections, so I guess we don’t need the local orientations after all.

Nevertheless, I think it would be handy if there was a way to get them in the non-loaded state. For my analysis of the loads, I needed to distinguish between the in-plane and out-of-plane bending moments. By default, these do not necessarily coincide with the local x- and y-axis, resulting in the tedious task of calculating the angle and rotating the beam elements with the ‘ElmRot’ parameter. If it would be possible to get the orientations of the local coordinate systems, this process could be fully automated.
On the other hand, I fully understand that a trade-off must be made here.

Anyways, thanks again for your help! My problem is solved.

Kind regards
Jelle