Hello,
I'm a student and I'm doing my master thesis work on modeling the seismic response of earth dams with abaqus, more specifically by doing dynamic non linear analysis. I did the analysis with a Mohr-Coulomb soil behavior and now I'm doing analysis with an UMAT non-linear material. After some attempts the model arrives at the end. Looking at the results, I have found these discontinuities that I can not interpret.
Someone can help me? Is it due to the use of triangular elements?
Thanks to everyone!
Be aware that what you see in a contour plot is extrapolated from integrations points. Especially in regions where plasticity occurs, this extrapolation may be off by a factor of 2 or more. Check the data at the integration points (tools->query) to see what happens in these regions.
See also this paper for more details on this:
Preprint How to get meaningful and correct results from your finite e...
A problem that usually occurs when working with large-scale problems is that you use "too many" elements!
Try reducing the number of elements by an order of magnitude and see what happens
In addition to what has been shared above, there is one more item you should be careful about: averaging regions. To plot contours, abaqus assigns elements to different averaging zones depending upon sectional/material properties, geometry, etc. If a node is shared by four elements belonging to different averaging zones, there'll be discontinuity in to contour plot. When in doubt, check integration point values.
Thanks to all for your reply.
I have tried to use triangular elements with 6 nodes instead of only three and the model works well.
- Badges
- Science topic