Hi Fluent Users,

Does anyone know how to regain control of Fluent in a journal file after an error has occured?

I am calling Fluent repeatedly from another piece of software, with a number of different meshes (running a genetic algorithm). Sometimes the mesh is rubbish, and Fluent either will not calculate, or hangs up after several iterations. In this case the journal file is aborted and Fluent just sits there doing nothing. Does anyone have any suggestions how to avoid this, or to detect it and do something about it? I've two ideas, but I don't know how to implement them in SCHEME:

1) if (fluent detects divergence) then (close fluent)

2) if (fluent is still open after 1000 seconds) then (close fluent)

Hoping some-one can help!! Hugo.

in scheme i think you have access to the actual time by simply typing (time). if it doesn't work you can also make system calls, like (system "batch_command")

i don't know how to detect divergence but you can maybe access the residuals and check if they're above a certain treshold

[FluidWarrior]

Hi,
you can catch errors in any programming language you want. "Try/Catch"
The Fundamentals are here:
https://www.researchgate.net/publica...t_and_ParaVIEW

Have a nice day!