Page 1 of 2
error in the simulation
Posted: Tue Apr 09, 2024 6:28 am
by simon007007
Hello,
I'm running a simulation but it crashes at some point and I can't figure out why...
From the histor.DAT file (attached) it seems going well but it stops and the error console says that there is a mismatch between the tag specified in the request and the tag present in the reparse point.
I'm using the latest SimVascular version and solver version on Windows.
Best regards,
Simon Molina
Re: error in the simulation
Posted: Tue Apr 09, 2024 1:27 pm
by davep
Hi Simon,
The mismatch between the tag specified in the request and the tag present in the reparse point. error is from the Windows OS, encounters a problem when writing a file, disk is full or corrupt maybe.
Cheers,
Dave
Re: error in the simulation
Posted: Wed Apr 10, 2024 6:10 am
by simon007007
Hello David,
thanks for your quick reply.
I will look into it.
Best regards,
Simon
Re: error in the simulation
Posted: Tue Apr 16, 2024 12:57 am
by simon007007
Hello,
I have emptied my disk and I don't have the same problem anymore.
However, I still have a problem
As you can see on the histor.DAT file, I have the NaN that appear at some point without further explanation and it doesn't even stop the simulation from running so I don't have an error message to work on...
Best regards,
Simon Molina
Re: error in the simulation
Posted: Tue Apr 16, 2024 1:54 pm
by davep
Hi Simon,
Is your simulation starting at time step 1? Make sure that there are no simulation files left over from your previous simulations, otherwise the solver will restart the simulation from those files.
Cheers,
Dave
Re: error in the simulation
Posted: Fri Apr 19, 2024 12:49 am
by simon007007
Hello,
yes I checked and deleted the files from the previous simulation so it starts at time step 1.
I refined the mesh so the simulation ran a bit longer but it still crashes suddenly (at 49%).
Should I continue to work on the mesh? I already have more than 1 million elements.
BR,
Simon
Re: error in the simulation
Posted: Fri Apr 19, 2024 3:25 pm
by davep
Hi Simon,
It's not clear what is going on here, the nonlinear residuals look good and then you suddenly get a NaN.
Upload you SV Simulations directory someplace I can download it and I'll have a look.
Cheers,
Dave
Re: error in the simulation
Posted: Fri Apr 19, 2024 3:55 pm
by davep
Hi Simon,
It's not clear what is going on here, the nonlinear residuals look good and then you suddenly get a NaN, doesn't seem to be a time step problem. What do your results look like before the failure?
Upload you SV Simulations directory someplace I can download it and I'll have a look.
Cheers,
Dave
Re: error in the simulation
Posted: Mon Apr 22, 2024 7:31 am
by simon007007
Hi,
the results look consistent in terms of velocity magnitude and direction, pressure.
Here's the google drive link of the whole project (I hope it works):
https://drive.google.com/file/d/1iqg6T3 ... drive_link
Best regards,
Simon
Re: error in the simulation
Posted: Tue Apr 23, 2024 1:53 pm
by davep
Hi Simon,
I looked at your simulation results, everything looks ok, no NaNs
Code: Select all
$ tail histor.dat
559 2.558E+05 3.826E-01 ( 7) 1.298E-02 1.060E-03 < 30364-13| 12> [ 0 - 0]
559 2.559E+05 2.098E-02 ( -5) 1.458E-03 5.822E-04 < 28316-14| 14> [ 0 - 0]
559 2.560E+05 3.690E-03 ( -12) 5.683E-04 3.003E-04 < 18549-14| 15> [ 0 - 0]
559 2.561E+05 1.255E-03 ( -17) 2.567E-04 1.381E-04 < 23291-13| 14> [ 0 - 0]
559 2.562E+05 5.004E-04 ( -21) 1.179E-04 6.316E-05 < 22820-14| 14> [ 0 - 0]
560 2.563E+05 3.796E-01 ( 7) 1.295E-02 1.060E-03 < 14040-13| 12> [ 0 - 0]
560 2.564E+05 2.075E-02 ( -5) 1.461E-03 5.799E-04 < 28316-14| 14> [ 0 - 0]
560 2.565E+05 3.642E-03 ( -12) 5.683E-04 2.985E-04 < 18549-14| 14> [ 0 - 0]
560 2.566E+05 1.240E-03 ( -17) 2.564E-04 1.369E-04 < 23291-13| 14> [ 0 - 0]
560 2.567E+05 4.946E-04 ( -21) 1.178E-04 6.275E-05 < 22820-14| 14> [ 0 - 0]
Are you running these simulations from SV or the command line? If you are running from SV then you should see a popup informing you that the simulation has failed and a Details button when selected might show some information about why the simulation failed.
Your mesh is a bit lumpy which will create some localized flow or stress artifacts but I don't think these would cause the solver to fail, you would see large residues I think.
If the simulation just stops then I am thinking that the OS is terminating it because it is exceeding some limit; disk space or runtime or some other resource.
Cheers,
Dave