Issue when reading .py file on Isight

[lilizam]

Hi, all.
I need to perform a Data Matching optimization with Isight and Abaqus, but the results from Abaqus need to be transformed into a datum CSYS to be compared with the experimental data.

I wrote a .py file that transforms the results to the new CSYS and then selects the nodal history I need and call those as new variables so I can recognise them when rereading the output file (when editing the Abaqus component).
I activate the options "use custom scripts" and "use python" and reread the .odb file, but nothing happens. I do not see the new variables and when running the optimization the values from the simulation won't match the ones from the experiments.

Any advice on this?

Someone already suggested meLink to write the transformed results into a new .odb file (which was very helpful!) but in that case, I would not know how to read this new file on the Isight environment. Will the software read all the possible files in the folder? Because I am not selecting the .py file either (only activating "use custom scripts" and "use python").

Thanks in advance!

 

[Mustaine3]

Afaik reads Isight the .odb, so if the transformed data are only temporary data and not stored in the .odb, then they are not available for Isight.
I would also say, that you have to use the srcipt to also store the data onto that .odb or create a new one. Or export the transformed field into an ASCII file via Report->Field Output. Maybe this means, that you can't do it all with one Isight component. Maybe you have to use one component to submit the analysis, then OS CMD to run the script and then another Abaqus component to read the results.

 

[lilizam]

Many thanks for your fast reply, Mustaine3!

The .py file is writing a .rpt file with the variables I request and it is not being generated when using Isight although it works when I read it with Abaqus.
I will try that out OS CMD to run the script and see if it works.

Cheers

 

[Mustaine3]

Opening the .odb is not the same as running a script through A/CAE and perform action on an .odb that is opened in A/CAE. The transformation and report options are coming from A/CAE.
So using an OS CMD to run A/CAE with a script (see noGui option) could be the way to mimic your manual work.