User login

Navigation

You are here

Question for post-processing after each increment and return it back to calculation (non-local damage)

Dear everyone:

  I am trying to use a nonlocal damage criteria to simulate the crack growth, which can reduce the mesh dependency.

In abaqus, VUMAT/UMAT is doing computation at integration point for each increment, but I want to do the post-processing after each increment to get the non-local state variable and return this value back to calculation.    

Such as taking average value of one state variable within a distance about the center of one element, then return it back to another state variable to this element.

 Does anyone know the way to post-process after each increment and return it back? Any user subroutines related to this?  If ABAQUS can not, how about LS-DYNA? 

 

  Thank you!

Johannes T.B. Overvelde's picture

Hey,

I have done this in Abaqus for the case I need the first buckling mode as an imperfection in post-buckling analysis. I have done this using the scripting possibility in Abaqus (instead of input files or GUI). You can use script to do a step, read input and change model etc. 

I wrote a one hour tutorial on Abaqus script to get started fast. You can find this here. In the same page I have given some example script files that incorparate the extraction of first buckling mode and editing the model for post-processing. I do have to mention that these files are only meant as an example of script, and are written to fully understand them without being familiar with script. If you have more questions on script, please ask!

Thank you so much!

 I know only a little about python script.   It that post-processing by python can only be done on the ODB file?  Is this true?  If it is, which means you need to output every increment,  I think this should be a very good idea for abaqus standard. 

However, I forgot to mention that my job is using abaqus explicit, which has thousands millions increment, and u know that I can't output all of them. 

 

Subscribe to Comments for "Question for post-processing after each increment and return it back to calculation (non-local damage)"

Recent comments

More comments

Syndicate

Subscribe to Syndicate