I encountered an issue while using the phase-field method to simulate fatigue. The cycles of fatigue to simulate can not be determined before analysis, so I adopt the restart analysis to continue the analysis if the extent of crack propagation doesn't reach the expected level. I adopt the dummy mesh which requires the number of SDVs to output the user element's results. It works well in first analysis, but when I continue the restart analysis, all the results of SDVs are zero while the NT11 results are normal. I try to print the value of nstatev in UMAT, and find it is zero in restart analysis. I don't know why it is and there is no useful information in Abaqus document or online. Has anyone encountered the same problem and how to deal with it ?
You can take the following approaches to resolve the issue:
https://caeassistant.com/product/advanced-umat-subroutine/
By employing these methods, I hope you can resolve your issue and successfully continue your fatigue analysis using the phase-field method.
Thanks for your advice. Actually, the key of the problem is the use of restart analysis. If I proceed the analysis directly instead of restart analysis, the UEL and UMAT all work well. So, could you give some more specific suggestions?
Jiaqi Wang Were you able to solve the problem? If you share your input file (both before and after restart), maybe I can find something wrong there.
I recently tested SDVs in restart analysis for coupled temperature displacement procedure and it worked fine.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Inorganic Materials
- Metals
- Alloys
- Steel