FDS used to study helideck turbulence (CAP437)

[Espen S Gåserud]

Hi,

 

Have anyone used FDS to study turbulence above a helideck, or do you think FDS could be used for this purpose?

 

CAP437 provides guidance and criteria for turbulence and states that “the limit on the standard deviation of the vertical airflow velocity of 1.75 m/s should not be exceeded”. I am new to LES turbulence modeling and have previously mostly worked with the k-epsilon model. There the standard deviation of the velocity (in all directions since it assumes isotropic turbulence) could be expressed as (2/3*k)^(1/2). How can this be done this using FDS/Pyrosim?

 

I know I need to do some reading on LES turbulence modelling. But before I pursue using FDS to study helideck turbulence any further, I thought I would ask for some expert advice.

Thanks in advance!

-Espen-

[Randy McDermott]

For this problem you have two major considerations: (1) external boundary conditions and (2) grid resolution.  Of these, I think (1) is the hardest to get right.  For (2), to estimate the turbulent kinetic energy (TKE) in LES you need to post process the results, subtract the mean and square the fluctuations about the mean.  Strictly speaking, this is not TKE because LES produces the filtered velocity field, not the true velocity field.  You need enough resolution so that the TKE is reasonably grid independent.  This will require a grid convergence study on TKE.  This is straight forward, and your biggest problem there is simply computing power.

But TKE is directly impacted by the boundary conditions.  Incoming turbulence is advected into the domain in reality.  How can you capture this in your simulation?  That is a big problem in LES of atmospheric flows, and there is a whole literature on turbulent boundary conditions.  You can read about the Synthetic Eddy Method (SEM) in the FDS user guide.  But in atmospheric flows the problem also involves the local stability of the atmosphere due to external heating---hot summer day on the tarmac much different then cool overcast conditions.  Usually some sort of measurement is needed to capture the condition and you can input this information into FDS using MEAN_FORCING and RAMPS.  See the LNG_Dispersion section of the validation guide.

--
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+unsubscribe@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/61875732-bfde-41d4-8bd6-486f2ee65e97%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Edward Liu]

Can the Reynolds_Stress or VEL_RMS be considered as output Quantity, including individual components in X, Y, Z directions?

It will be useful to study the turbulence.

On Friday, April 21, 2017 at 8:43:33 AM UTC-5, Randy McDermott wrote:

For this problem you have two major considerations: (1) external boundary conditions and (2) grid resolution.  Of these, I think (1) is the hardest to get right.  For (2), to estimate the turbulent kinetic energy (TKE) in LES you need to post process the results, subtract the mean and square the fluctuations about the mean.  Strictly speaking, this is not TKE because LES produces the filtered velocity field, not the true velocity field.  You need enough resolution so that the TKE is reasonably grid independent.  This will require a grid convergence study on TKE.  This is straight forward, and your biggest problem there is simply computing power.

But TKE is directly impacted by the boundary conditions.  Incoming turbulence is advected into the domain in reality.  How can you capture this in your simulation?  That is a big problem in LES of atmospheric flows, and there is a whole literature on turbulent boundary conditions.  You can read about the Synthetic Eddy Method (SEM) in the FDS user guide.  But in atmospheric flows the problem also involves the local stability of the atmosphere due to external heating---hot summer day on the tarmac much different then cool overcast conditions.  Usually some sort of measurement is needed to capture the condition and you can input this information into FDS using MEAN_FORCING and RAMPS.  See the LNG_Dispersion section of the validation guide.

On Fri, Apr 21, 2017 at 12:24 AM, Espen S Gåserud <esp...@gmail.com> wrote:

Hi,

 

Have anyone used FDS to study turbulence above a helideck, or do you think FDS could be used for this purpose?

 

CAP437 provides guidance and criteria for turbulence and states that “the limit on the standard deviation of the vertical airflow velocity of 1.75 m/s should not be exceeded”. I am new to LES turbulence modeling and have previously mostly worked with the k-epsilon model. There the standard deviation of the velocity (in all directions since it assumes isotropic turbulence) could be expressed as (2/3*k)^(1/2). How can this be done this using FDS/Pyrosim?

 

I know I need to do some reading on LES turbulence modelling. But before I pursue using FDS to study helideck turbulence any further, I thought I would ask for some expert advice.

Thanks in advance!

-Espen-

--
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]

No. Just use a 3D slice of velocity components. Then compute yourself using sl3d_read.m

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/37b352eb-a713-4072-9492-85626970ca8d%40googlegroups.com.

--

Sent from my iPhone

[Edward Liu]

The post-processed TKE could not be visualized in SmokeView.

Is there another way to output or calculate it and show it in Smokeview?

On Tuesday, February 4, 2020 at 5:26:36 PM UTC-6, Randy McDermott wrote:

No. Just use a 3D slice of velocity components. Then compute yourself using sl3d_read.m

To unsubscribe from this group and stop receiving emails from it, send an email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/37b352eb-a713-4072-9492-85626970ca8d%40googlegroups.com.

[dr_jfloyd]

The format for smokeview files in in the FDS User's Guide. Write your processed data in the smokeview format and edit your smv file (it is ASCII text) so that smokeview knows your new file is there.