Solutions not saved when using batch file

[sigurd.naess]

Hi!

Earlier today I tried to run  two non-linear static analyses using a batchfile (as described in the link below)

The first analysis has completed and was saved, but the second was not.

The second analyses was to find the plastic collapse load under pressure i.e. analysis is run until it cannot converge. Is this the reason the solution does not save to .FNO? is there a workaround?

Thanks!

[John_Holtz]

Hi @sigurd.naess 

What does the log file indicate about the second analysis? Did it complete? Was there an error before any results were output? (That is, before the FNO file was created.) 

What are the extensions for the files that were created?

Is the problem because you ran two analyses in the batch file? If you run the second analysis by itself, does it work?

[sigurd.naess]

Hi @John_Holtz !

I've done some additional testing today. I have 4 different geometries that i'm analyzing

First batch run consists of:

A) 4 x heat transfer (linear)

B) 4 x non-linear static (@ working pressure) includes the temperature gradients from the previous analyses

C) 4 x non-linear static (@ arbitrary high pressure load that will cause plastic collapse ) includes the temperature gradients from the previous analyses

The .FNO is saved for A) and B), but not for C)

Second batch run consists of only:

C) 4 x non-linear static (@ arbitrary high pressure load that will cause plastic collapse ) includes the temperature gradients from the previous analyses

The .FNO is not saved for this case either.

I've included TIMEOUT 10 between each analysis

Each non-linear static are set to NOT save incremental output (perhaps this shoul be on?) 

The included picture displays all the files for the first geometry:

 x5mpqn80j --> heat transfere

hfzg4c5b3 --> non-linear static @ working pressure

3datin2k1 --> non-linear static to plastic collapse

I was hoping to solve ten cases of plastic collapse (as a batch) during the weekend. Great if you can confirm if this is possible or not.

Regards
Sigurd

[John_Holtz]

Hi @sigurd.naess 

I think that you have solved your own case.😀

If you tell the software to only output results at 100% load, and if the analysis fails before reaching 100%, there will be no results. (Your analysis failed with a "FATAL ERROR E5076: MAXIMUM NUMBER OF BISECTIONS PERMITTED REACHED" at 81% complete.)

You want to set the "Nonlinear Setup > Intermediate Output" to "On". This will save the results at each completed step.

[sigurd.naess]

😅😅

I will give it a shot right away!

Thank you!

[sigurd.naess]

Hi again @John_Holtz 

I just completed a single analysis from batch file and still it did not save the .FNO even though I set  Intermediate Output" to "On" --> see attached picture 😥

Just to be sure, the .FNO loads the results to the Nastran interface right? I want to be able to view the plots

Regards
Sigurd 

[John_Holtz · Accepted solution]

What? I can't believe it. ☹️

I am hoping that the one thing that I did not mention previously is the solution. (I was doing my test in the interface, so it did not occur to me to mention this. 🙄) After changing the option to output the intermediate results, you would need to regenerate the Nastran file. You do this by right-click in the model tree and choosing "Generate Nastran File" (or starting the analysis in the interface will regenerate the Nastran file). If you do not regenerate the Nastran file, then the change made in the interface it not transferred to the Nastran file.

If that is not the solution, please attach the .nas file to the forum.

[sigurd.naess]

Hi @John_Holtz 

It worked!!!! 😂

In summary:

- Intermediate Output = on

- utdate .NAS file to capture this

- run batch file 

Regards
Sigurd