Hi,

Abaqus does have the ability you need. You only have to request a Field Output or a History output for the nodes you want to evaluate. There you can also specify how many points in time should be saved. In the pre-processing stage ypou only need to save the XY-Data and write a Report file. This is a text file that can easily be imported in Matlab and further processed.

Best,

@Sascha Duczek ... Thanks a lot sir for your detail explaination , it helps me . Thank you .

As an addition to what +sascha Duczek said - you can also do this by writing a small python script. This is especially useful if you have to work on a lot of configurations. A simple example for extracting history output results is provided in the Abaqus Scripting manual.

One more thing - if you need history data, then it is more advisable to save the data you need as a History variable rather than a Field variable.

Another solution could be the User Subroutine URDFIL. Pros: It allows you to output whatever you want into a '.dat' file so as to read it from Matlab. Cons: you need FORTRAN knowledge.

+Miguel A. Ariza-Gracia - doesn't URDFIL actually write to the fil file?

Personally, I would avoid writing to the dat file - if one is working with large amounts of data, it just blows up since the *.dat file is ASCII. I prefer the odb file since it is binary and only the relevant data can then be extracted.

Arun, yes indeed. I also see your point, but in my experience I've never experienced probls while writing data (a large amount, believe me). I also used python scripts but I found it quite slow. And I also used odb, but I use to work with some clusters and moving heavy odb files is not very useful!

Cheers

Thanks you Arun Prakash and Miguel for sharing such a useful knowledge, i appreciate it very much !