CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Numercial Simulation on Contaminant Distribution (

Apple L S Chan November 2, 1998 04:30

Numercial Simulation on Contaminant Distribution
I am a PhD student and now conducting CFD simulation on indoor contaminant distribution. First of all, a converged solution of velocity and temperature distribution has been obtained by a CFD program. Then the transport equation for contaminant concentration is going to be solved.

My question is that how I can get the value of contaminant concentration at varios time interval, e.g. each minute after the source began to generate contaminant. Should I run the simulation for 1 minute and then stop it and get the simulated concentration? And then resume the simulation for another 1 minute?

Could anyone please help me and give me some advices? Thank you!

Joern Beilke November 2, 1998 12:06

Re: Numercial Simulation on Contaminant Distribution
Run a transient solution and start injecting the contaminant at a time greater t0, when your flow field seems to be ok.

Heinz Wilkening November 4, 1998 05:31

Re: Numercial Simulation on Contaminant Distribution

I am not sure if you could realy decouple the problem of distribution and fluid flow as you suggest. I have done quite a lot containt distribution calculations after a release of Gases cause by an accient scenario.

If the source is not realy passiv - means very low volume flow rates, same thermodynamic state such as pressure density and temperature and ... - then there will be a strong oupling between the source and the mean flow path. This is enhanced if heat transfer to structure plays a important role. Then natural convection loops can change dramatically and influence the overall distribution very much.

I would suggest to calculate the CFD-Part and the Distribution together at the same time in a trasient way.

You can contact me, if you like to know some referneces, but I hve to search for this a little, so I can't tell right now.


Shia-Hui Peng November 5, 1998 05:11

Re: Numercial Simulation on Contaminant Distribution
Indoor contaminant dispersion in many cases can indeed be (and actually has been in a lot of previous studies) treated as being of a passive type in numerical simulations since the generated contaminant usually possesses nearly same physical properties as room air (e.g. some VOCs released from building material etc.). The mass transport equation can thus be solved separately being decoupled from the momentum and thermal energy equations.

There exist also many cases where these equations must be solved in coupling with each other. This is true, for instance, when the contaminant is dynamically released from a source and, in particular, when the contaminant is a type of aerosal and imposes siginificant influences on the flow field. For particulate contaminants, a stochastic treatment of the particle-turbulence interaction has often been employed by tracking the trajectories of particles of representive sizes.

As regards how to get the computed result at various time intervals, I cannot see any problem with it. It is certainly not necessary to terminate the simulation for saving and then to restart it again. A simple way is to save the concentration values in a file DURING the calculation. If one is only interested in the values (changing with time) at a few of points, the storage capacity required by such a file should be acceptable.

theja October 24, 2011 07:01

particle and fluid simulation
hai shia

i am working on the simulation of dust particles in room environment,
since i am a new user of openfoam i need your help on choosing the type of solver and guidance in solving the problem

All times are GMT -4. The time now is 14:23.