Abaqus Explicit - Error in restart analysis

[Maria Kxr]

Hello everyone,


I am trying to run a restart analysis and I get the following error"

"A multi-domain restart analysis has modified the boundary conditions at one or more nodes that are either involved in a kinematic constraint or are part of a rigid body that is involved in a kinematic constraint. Because the domain decomposition is based only on the features specified in the original analysis and steps defined therein, modifying the boundary conditions in the restart analysis would invalidate the original domain decomposition. Please check the boundary conditions at nodes/rigid bodies involved in kinematic constraints. A dummy step may be introduced in the original analysis with the same set of boundary conditions on these nodes as defined in the restart step, to prevent this error."


I am not changing anything in my BCs, I am just copying and pasting my model. In the original analysis I run step-1, and in the restart analysis I want to begin from step-1 and run step-2.

I cannot find anything for this error in the documentation. Does anyone know what it means and how I can fix it?


Thank you in advance.

 

[FEA way]

Are you defining the restart analysis in Abaqus/CAE or in an input file ? The procedure differs depending on that.

 

[Maria Kxr]

I am using Abaqus/CAE currently.

 

[FEA way]

Then you should copy the original model without changing anything at or before the restart location. You can only add sets, amplitude definitions and new steps (without changing the previous ones). To make sure nothing is changed even accidentally, it might be better to perform a restart from an input file. If you follow this advice and still get the error then I suggest doing what the error message suggests - adding a dummy analysis step with no changes and then following it with a new analysis step which actually adds something to the analysis history.

 

[Maria Kxr]

So what I had so far was:

Analysis 1: Step 1, Step 2 -> Terminate analysis at step 1
Analysis 2: same steps (1 & 2) -> Restart request from step 1 of previous analysis

Is this the same as the procedure you described, or am I missing something?

 

[FEA way]

So you are trying to restart an analysis that was terminated prematurely, not add new steps to already completed simulation ? Then look for the recovery procedure in Abaqus/Explicit.