Simulation Error - [forrt1: severe (157): Program Exception - access violation]

[Daniel Pau]

Hi All,

One of my FDS input file has a total of 1,536,024 meshes and the PC that I am using to run that input file has 16 GB of ram. The input file has a total of around 420+ devices (mainly consisted of devices to measure visibility, FED, temperature, radiant heat flux gas).

The input file runs perfectly fine without the devices. However, when I incorporated all the devices into the input file, the attached error (see attachment) occurred.

I am wondering if it is due to insufficient memory to carry out the operation? Can anyone confirm that? Do devices take up memory?

I have another input file which has a total of 5,108,376 meshes also has the same problem (runs fine without devices). 1.5 million meshes are significantly less than 5 million meshes, but yet they have the same problem. If it is truly a problem with insufficient memory, then those devices must have taken up loads of memory which I don't believe is the case (correct me if I'm wrong). This leads me to think that the problem might not be anything to do with memory but something else perhaps. 

It will be much appreciate if someone can help shed some lights on this problem. Many thanks in advance.

Regards

Daniel

[Mohamed Assal]

Hello Daniel,

1,536,024 meshes you mean Cells.!

So i don't think so that your problem is related to memory, just to
eliminate this issue can you please share the performances of your
computer, CPU...etc.

Can you share also your input.

Cheers Mohammed.

2015-01-18 13:09 UTC+01:00, Daniel Pau <daniel...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups
> "FDS and Smokeview Discussions" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to fds-smv+u...@googlegroups.com.
> To post to this group, send email to fds...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/fds-smv/d1691489-abda-4d10-817f-7d81f7103c58%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>

[Daniel Pau]

Hi Mohammed,

Thanks for your reply.

Apologies, yes I meant cells.

Performance of my computer and the input file containing 1.5 million cells is attached.

Could it be due to the number of devices I have? Is there a limit on the number of devices?

Thanks

Daniel

[Glenn Forney]

do you mean 1.5 million meshes or grid cells.  if meshes that is way too many.  smokeview wouldn't be able to view such a case.  simplify your case and scale up gradually.

--

[dr_jfloyd]

If it runs without the devices, then most likely you have an error that we are not trapping in input processing.  Take the file without devices and add the devices back in small groups.  When the input fails, the error will be in the last group added.  You can then add that group in smaller sections, etc. until you find the device input causing the problem.

[Daniel Pau]

Hi GF,

My apologies, I am supposed to say grid cells. I am still fairly new to FDS and computational fluid dynamics in general and struggle with terminology at times. Sorry for the confusion there.

Regards

Daniel

[Daniel Pau]

Hi Dr J Floyd

Thanks for the suggestion. I shall try that and see what happens. 

Regards,

Daniel

[Daniel Pau]

The problem seems to be caused by the radiative heat flux gas devices. The error showed up as soon as a radiative heat flux gas device is added. I have tried different ways of writing the command for the device and all of them came up with the same error. As an example, I have tried the following:

&DEVC XYZ=-0.825,1.2,2.0, QUANTITY='RADIATIVE HEAT FLUX GAS', ID='HF_(-0.825-1.2-2.0)' /

OR

&DEVC XYZ=-0.825,1.2,2.0, QUANTITY='RADIATIVE HEAT FLUX GAS', ID='HF_(-0.825-1.2-2.0)', ORIENTATION=0,0,-1 /

OR

&DEVC XYZ=-0.825,1.2,2.0, QUANTITY='RADIATIVE HEAT FLUX GAS', ID='HF_(-0.825-1.2-2.0)', IOR=-3 /

I have also ran some test cases with the above commands and they all seem to run fine. Not sure what else can I do here... Any help will be much appreciated.

On Monday, January 19, 2015 at 2:31:47 AM UTC+13, dr_jfloyd wrote:

[Mohamed Assal]

Dear Daniel

Try script like this daniel

&DEVC XYZ= 2.6, 2.3, 0.0, QUANTITY='RADIATIVE HEAT FLUX', IOR=3, ID='RAD_FLUX' /

Cheers Mohammed

2015-01-19 0:35 UTC+01:00, Daniel Pau <daniel...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups
> "FDS and Smokeview Discussions" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to fds-smv+u...@googlegroups.com.
> To post to this group, send email to fds...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/fds-smv/be955fa6-65c2-453d-a370-bf4939554d92%40googlegroups.com.

[Daniel Pau]

Hi Mohamed,

I have tried the command line that you have suggested below and it came up with an error asking for the device to be reposition as FDS cannot determine which boundary cell to assign. I remember reading in the manual somewhere that the device measuring "RADIATIVE HEAT FLUX" will have to be assigned on solid surface but for the purpose of my study, I am looking to monitor the radiative heat flux at a floating point 2 m above floor, hence why I used "RADIATIVE HEAT FLUX GAS". 

All in all, it still doesn't work but thanks for the suggestion.

Regards

Daniel 

[dr_jfloyd]

Make sure that you are putting the device in a gas cell.  You should be using the ORIENTATION version of the input. If you are putting it in a gas cell, then simplify your input file as much as possible while still getting the error and post it to the issue tracker.

[Daniel Pau]

It appears that the problem is caused by evac commands. I created a very simple test file comprising of a box with some radiative heat flux devices and few agents exiting the box (see attached). The same error came up as I tried to run the test file. I'll post this to the issue tracker.

[TimoK]

Well, I'm back from winter holidays. This is already in the issue tracker, issue 2330.

If you need the DEVC output, you need to do the fire simulation without the evacuation
to get the output. Then take the fire output away and run fire+evacuation calculation.

You can edit the CHID.smv files or copy the output files so that you can see
humans and fire related output in Smokeview at the same time. But you need
not to do this, because the DEVC output seems to be the problem and this
you are not going to see  in Smokeview.

Timo