input the displacement field as a boundary conditon in abaqus

[Yu12366]

thread1630-493737
Hi,

I want to input the displacement of nodes calculated from the last simulation as a boundary condition for the new simulation (same geometry and mesh).

I have tried to create the mapped field with Odb mesh, following the instructions from the User's Manual. However, after I created this in 'Analytical Field', I could not select it from 'Distribution - Displacement/Rotation - BCs' (only Unifrom and User-defined existed).

Do you have any ideas why this was and how I should input the displacement BC?

Yours,
Yu

 

[rb1957]

do you want to restart the analysis ? have large displacement (non linear geometry) on ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Yu12366]

Yes. I will use this displacement as BC and re-run the explicit simulation with large displacement on.

 

[FEA way]

Check this thread:

https://www.eng-tips.com/viewthread.cfm?qid=493737

 

[Yu12366]

I was following the thread you provided. However, as I have mentioned: after I created a mapped field with Odb mesh in 'Analytical Field', I could not select it from 'Distribution - Displacement/Rotation - BCs' (only Unifrom and User-defined existed).

 

[rb1957]

"surely" ABAQUS can restart an analysis without inputting a set of deformations ?

or are you trying to "reverse engineer" the loads ? "If I know the deformations then applying them as a BC would tell me the loads that created them"?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Yu12366]

Yes. I would like to get something similar to what you described. Basically, I would like to apply a load to the initial geometry to get deformations, then this deformation field will be subtracted from the initial geometry so that I can get the geometric models before the load is applied.

 

[rb1957]

sorry, but WTF ? what you're saying is you have the undeformed, unloaded model; you've got a set of loads to apply to it, so it deforms. You then want to apply these deformations to the deformed model to get back to the original model.

1) I don't think this is going to work.
2) Y, oh Y ??

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Yu12366]

I would apply a set of loads to the original model (G0) to get deformed geometry (G1), then the displacement field (D = G1-G0) will be subtracted from the original geometry (G0), not the deformed geometry (G1) as you have mentioned.

The reason is that the original model (G0) was reconstructed from the medical image, and it is already under a pressure load and I need to get its unloaded shape. We assumed that the deformation field (D) from unloaded geometry to G0 would be the same as from G0 to G1 under the same pressure load, and so knowing D and substracted it from G0 would give me the unloaded model.

Still, the question is (1) after I created a mapped field with Odb mesh in 'Analytical Field', I could not select it from 'Distribution - Displacement/Rotation - BCs' (only Unifrom and User-defined existed). Why would this happen?

(2) Are there any other ways to map the deformation field?

 

[rb1957]

ok, so you know G1 (the deformed structure under "pressure") and you know G0 (the undeformed structure).

I don't get "so knowing D and substracted it from G0 would give me the unloaded model" ... isn't the unloaded structure G0 ?

Or are we getting terminology mixed up ? 'cause you say "the original model (G0) was reconstructed from the medical image, and it is already under a pressure load" but also say "apply a set of loads to the original model (G0) to get deformed geometry (G1)".

Let's settle on terminology ...
Gd is the deformed structure you have, under load.
Do you know Gu ... the structure without load ? but then this seems to be what you're trying to determine ??

I can't help with ABAQUS, and don't understand what you are trying to achieve. We can resolve this by you saying "I know what I'm trying to do, and just need help with doing it in ABAQUS."

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Yu12366]

Sorry to confuse you and I know it would be a little bit difficult to explain.

It is quite common in our field so yes, I definitely know what I'm trying to do and please I just need help with doing it in Abaqus.

 

[FEA way]

Unfortunately, you may have to use Python scripting or Fortran subroutine (DISP) to achieve this. Only analytical mapped fields are supported in Abaqus/CAE for displacement BCs.

 

[Yu12366]

Thanks for your reply. It seems that the guy in the following thread (as you have mentioned) successfully implemented the displacement field using the Mapped Field so I am wondering what is the difference between us?

https://www.eng-tips.com/viewthread.cfm?qid=493737

 

[FEA way]

I don't know what the OP of that thread managed to achieve (maybe he just found a way to create the mapped field from the BC definition window without actually assigning it) but only expression fields can be used for displacement BCs.

 

[rb1957]

maybe he fibbed ?

or likely, he found another way to do whatever it was and didn't get back ??

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Yu12366]

Thanks both!

I have managed to do it by exporting the deformation field first and then making it as an input file that can be read with the model inp.file.

Thanks for your help and suggestions!