[khariel]

I'm wondering what would be the best way to save the results of my simulations on CFX. I have around 50-60 'quick' simulations to run, where I'm evaluating the behavior of my systemby changing a single parameter.

I still haven't figured out yet how and what I'm going to plot from the results I get, so I just wanted to store them for now. I'm using Workbench.

Thanks in advance,

[ghorrocks]

If the output boils down to a single number then you can do it as a parametric thing inside workbench. You can also run this inside a script/batch file and do the same thing from the command line.

[khariel]

Originally Posted by ghorrocks

If the output boils down to a single number then you can do it as a parametric thing inside workbench. You can also run this inside a script/batch file and do the same thing from the command line.

Hello Glenn, thanks for the reply.
What I was looking for was more in terms of saving all the result variables (velocities, pressure, mass fractions, etc) for each simulation in a way that I'd be able to, at any given time, open these results in CFD-Post and visualize them.

[ghorrocks]

The results file contains all these variables by default. You don't need to do anything to have these variables available.

[khariel]

Originally Posted by ghorrocks

The results file contains all these variables by default. You don't need to do anything to have these variables available.

So if inside the same 'analysis' on Workbench I run a simulation, then later I change something and run another simulation, will both (and all subsequent) results be available at the respective .res file?

[ghorrocks]

In workbench - it depends on how you set it up. If you edit a workbench setup and rerun then the old run will not be kept.

Personally I like to have more control over what is going on than workbench allows. So I would run this as stand-alone CFX using a batch file to drive it from the command line. Then it is easy to make sure all files are kept and organised.

[khariel]

Originally Posted by ghorrocks

In workbench - it depends on how you set it up. If you edit a workbench setup and rerun then the old run will not be kept.

Personally I like to have more control over what is going on than workbench allows. So I would run this as stand-alone CFX using a batch file to drive it from the command line. Then it is easy to make sure all files are kept and organised.

Where can I read more about that to learn how to use the stand-alone version with the command line?

[ghorrocks]

You access it from the Fluid Dynamics/CFX Launcher on the start menu.

The CFX documentation has the command line parameters under "Starting CFX from the command line".

[khariel]

Originally Posted by ghorrocks

You access it from the Fluid Dynamics/CFX Launcher on the start menu.

The CFX documentation has the command line parameters under "Starting CFX from the command line".

I'll give it a go. Thanks a lot once again for your patience, Glenn.

[Steffen595]

all the results are in a subfolder inside a folder with the same name as the project. So you an open them in CFD post. And it saves a text file with your parameters. This is only, if you pick up your last iteration, change parameters in Pre, and run another calculation. Guess there is some setting. Because Fluent overrides by default, whereas CFX just keeps making new results files.

[Antanas]

Originally Posted by khariel

So if inside the same 'analysis' on Workbench I run a simulation, then later I change something and run another simulation, will both (and all subsequent) results be available at the respective .res file?

If you open properties of Solution cell and set Multi-configuration Post Processor File Load Options to one of Complete Histories then it will keep res and out files for every update of Solution cell unless you clear its generated data.

[Steffen595]

and then in folder (projectname)_files\dp0\Cfx\Cfx

[khariel]

Originally Posted by Antanas

If you open properties of Solution cell and set Multi-configuration Post Processor File Load Options to one of Complete Histories then it will keep res and out files for every update of Solution cell unless you clear its generated data.

Oh, interesting. I'll give this a try too. Thanks.
My only concern though is that the results files start to get very large.

By the way, if I run a simulation, then change some parameter (say, inlet velocity), and run it again without clearing generated data, how does it affect my second simulation? because I'm seeing my RMS values already starting very low, and my monitor point has the same value as the previous simulation.

[Antanas]

Originally Posted by khariel

Oh, interesting. I'll give this a try too. Thanks.
My only concern though is that the results files start to get very large.

By the way, if I run a simulation, then change some parameter (say, inlet velocity), and run it again without clearing generated data, how does it affect my second simulation? because I'm seeing my RMS values already starting very low, and my monitor point has the same value as the previous simulation.

CFX will create files for every update, so they won't grow. RMS starting very low for two reasons:
1. CFX continues history from prev. run. In general residuals will jump up after 1st step.
2. It uses results of prev. run as initial conditions. You may change this by modification Initialization option in properties of Solution cell.

[Steffen595]

and the .out file is a text file that has all your settings / inputs.
The .res file size depends on the size of your mesh and how many variables you calculate, and it makes a new one for every run.
In Post you can open each .res file individually. Its a good idea, if you save your settings, but it asks you anyway.
If you only vary one variable slightly or use that calculation for something similar, then its not bat that it uses your previous values for initialisation, saves you some time.