Spatial and Temporal Statistics in FDS

756 views
Skip to first unread message

Kevin

unread,
Feb 6, 2019, 11:36:38 AM2/6/19
to FDS and Smokeview Discussions
Another new feature in FDS 6.7.1 is the enhancement of the STATISTICS parameter for output data. There are two things to keep in mind, and more details can be found in the User's Guide.

1. In past versions of FDS, if you invoked, say, STATISTICS='VOLUME INTEGRAL', the bounds of the integration had to lie completely within one mesh. This is no longer necessary. All spatial statistics ought to work on multiple meshes.

2. The parameter STATISTICS has been replaced by SPATIAL_STATISTIC and TEMPORAL_STATISTIC (but STATISTICS will still work). The reason for this change is that now you can combine, for example, temporal and spatial averaging. 

Use this discussion thread for questions and comments on this new feature.

fde

unread,
Feb 8, 2019, 5:51:05 AM2/8/19
to FDS and Smokeview Discussions
I have an example in which I tried using SPATIAL_STATISTIC='SURFACE INTEGRAL' over multiple meshes. However, there seems to be a problem.

Both devices output the same result.


  
&HEAD CHID='2_mesh'/
&TIME T_END=20.0/
&DUMP  DT_DEVC=1,   DT_PART=1.0, DT_SLCF=1.0/
&MISC ALLOW_UNDERSIDE_PARTICLES=.TRUE. NOISE=.FALSE. PARTICLE_CFL=.TRUE. /
&TIME DT=0.005/
&RADI RADIATION=.FALSE./
 
&MESH ID='1',   IJK=20,10,10, XB=0,2,0,1,0,1/, MPI_PROCESS=0/
&MESH ID='2',   IJK=20,10,10, XB=2,4,0,1,0,1/, MPI_PROCESS=1/
 
&DEVC ID='Nozzle+x', PROP_ID='K', XYZ=0.5,0.5,0.7, ORIENTATION=1.0,0,0, QUANTITY='TIME', SETPOINT=0.0/

&DEVC ID='AMPUA_Ground_2m', XB=0, 2, 0, 1, 0, 0,  PART_ID='Water',  QUANTITY='AMPUA',  SPATIAL_STATISTIC='SURFACE INTEGRAL'  /
&DEVC ID='AMPUA_Ground_4m', XB=0, 4, 0, 1, 0, 0,  PART_ID='Water',  QUANTITY='AMPUA',  SPATIAL_STATISTIC='SURFACE INTEGRAL'  /

 
&SPEC ID='Water_SPEC', FORMULA='H2O', DENSITY_LIQUID=1000.0, SPECIFIC_HEAT_LIQUID=4.184, VAPORIZATION_TEMPERATURE=100.0, MELTING_TEMPERATURE=0.0, HEAT_OF_VAPORIZATION=2259.0/

&PART ID='Water',
      SPEC_ID='Water_SPEC',
      DIAMETER=300.0,
      DISTRIBUTION='ROSIN-RAMMLER',
      RGB=0,204,0,
      AGE=60.0,
      SAMPLING_FACTOR=1
   QUANTITIES='PARTICLE AGE', 'PARTICLE VELOCITY'/

 
&PROP ID='K',
      OFFSET=0.05
      PART_ID='Water',
      PARTICLES_PER_SECOND=5000,
      FLOW_RATE=12
      FLOW_TAU=.5,
      ORIFICE_DIAMETER=4E-3,
      SPRAY_ANGLE=0.0,25.0/
 
 
&VENT ID='Vent', SURF_ID='OPEN', XB=4,4,0,1,0,1/  
&BNDF QUANTITY='AMPUA', PART_ID='Water'/
&TAIL /

dr_jfloyd

unread,
Feb 8, 2019, 8:14:00 AM2/8/19
to FDS and Smokeview Discussions
Can you submit this case to the issue tracker? Thanks.

Matt

unread,
Apr 29, 2019, 6:07:53 PM4/29/19
to FDS and Smokeview Discussions
That is fantastic news!  A macro method of dissecting meshes is now much simpler....  I use Statistics all the time and believe its the best way of determining pass/fail criteria.  Statistic 'boxes' crossing meshes shouldn't be a problem now!

Many thanks

Matt. 

Vishwa

unread,
May 10, 2020, 11:21:18 AM5/10/20
to FDS and Smokeview Discussions
Mr.Kevin,
Is there a method to take spatially distributed temporal output quantities in SLCF namelist group? 

Randy McDermott

unread,
May 10, 2020, 1:09:17 PM5/10/20
to FDS and Smokeview Discussions

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/c32ff906-6b5a-43bc-8f02-cbbe398bd603%40googlegroups.com.

Vishwa

unread,
May 10, 2020, 1:29:03 PM5/10/20
to FDS and Smokeview Discussions
I used it but still when taking SLCF outputs , I can't take temporal values. I can only get values for a averaged time period isn't it?
I herewith, attached the command prompt snapshot when using fds2ascii.


On Sunday, May 10, 2020 at 10:39:17 PM UTC+5:30, Randy McDermott wrote:
On Sun, May 10, 2020 at 11:21 AM Vishwa <vish...@gmail.com> wrote:
Mr.Kevin,
Is there a method to take spatially distributed temporal output quantities in SLCF namelist group? 



On Wednesday, February 6, 2019 at 10:06:38 PM UTC+5:30, Kevin wrote:
Another new feature in FDS 6.7.1 is the enhancement of the STATISTICS parameter for output data. There are two things to keep in mind, and more details can be found in the User's Guide.

1. In past versions of FDS, if you invoked, say, STATISTICS='VOLUME INTEGRAL', the bounds of the integration had to lie completely within one mesh. This is no longer necessary. All spatial statistics ought to work on multiple meshes.

2. The parameter STATISTICS has been replaced by SPATIAL_STATISTIC and TEMPORAL_STATISTIC (but STATISTICS will still work). The reason for this change is that now you can combine, for example, temporal and spatial averaging. 

Use this discussion thread for questions and comments on this new feature.

--
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...@googlegroups.com.
1s.PNG

Vishwa

unread,
May 10, 2020, 1:33:31 PM5/10/20
to FDS and Smokeview Discussions
Mr.Randy,

Is there a specific method to get spatially distributed temporal values?

P.S. In DEVC namelist, it is clearly mentioned in the user guide that we can get temporal output values. Since I need both spatially and temporal values it can't be used.

Thank you


On Sunday, May 10, 2020 at 10:39:17 PM UTC+5:30, Randy McDermott wrote:
On Sun, May 10, 2020 at 11:21 AM Vishwa <vish...@gmail.com> wrote:
Mr.Kevin,
Is there a method to take spatially distributed temporal output quantities in SLCF namelist group? 



On Wednesday, February 6, 2019 at 10:06:38 PM UTC+5:30, Kevin wrote:
Another new feature in FDS 6.7.1 is the enhancement of the STATISTICS parameter for output data. There are two things to keep in mind, and more details can be found in the User's Guide.

1. In past versions of FDS, if you invoked, say, STATISTICS='VOLUME INTEGRAL', the bounds of the integration had to lie completely within one mesh. This is no longer necessary. All spatial statistics ought to work on multiple meshes.

2. The parameter STATISTICS has been replaced by SPATIAL_STATISTIC and TEMPORAL_STATISTIC (but STATISTICS will still work). The reason for this change is that now you can combine, for example, temporal and spatial averaging. 

Use this discussion thread for questions and comments on this new feature.

--
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...@googlegroups.com.

Randy McDermott

unread,
May 10, 2020, 1:33:44 PM5/10/20
to FDS and Smokeview Discussions
fds2ascii will give you one time average over the interval you specify.

If you want to do more than that, slread.m returns an array with all the data in the slice file.  You can then do whatever you want with it.  But you have to be Matlab savvy.  If you prefer Python, a version also exists:


There is not much more we can do to give you access to the slice data.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/ff3e95a8-b6d8-4f7b-9ff9-5c0ea9862190%40googlegroups.com.

Randy McDermott

unread,
May 10, 2020, 1:34:22 PM5/10/20
to FDS and Smokeview Discussions
Use slread.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/28883506-d232-4d99-a5cf-be12c7ce1638%40googlegroups.com.

Vishwa

unread,
May 10, 2020, 1:40:05 PM5/10/20
to FDS and Smokeview Discussions
Thank you very much. 
I will refer slread
Use slread.

Reply all
Reply to author
Forward
0 new messages