[SQAK]

Hello,

I am new to CFD. I work at a petrochemicals complex. My function requires me to make a judgement on which petrochemical projects will need a CFD modelling study in additional to normal 2D consequence modelling.

I know congestion is the main factor but how to define qualitatively or quantitatively what is congested and what is not.

Any thumb rules docs?

[SQAK]

Can someone help in deciding what could be considered "congested" to require CFD over and above normal consequence modelling?

[sbaffini]

I am afraid you are up to something very specific and not directly related to CFD, which might be out of reach for those not working in your specific field.

For example, I have no idea what congestion or consequence modeling refer to.

This means you will probably only get help from someone who exactly knows what you're talking about... which might or not be around in these days.

[SQAK]

Originally Posted by sbaffini

I am afraid you are up to something very specific and not directly related to CFD, which might be out of reach for those not working in your specific field.

For example, I have no idea what congestion or consequence modeling refer to.

This means you will probably only get help from someone who exactly knows what you're talking about... which might or not be around in these days.

Ok. May be I will try in software specific forum like ANSYS or FLACS. Thanks!

[hokhay]

https://www.dnvgl.com/Images/Introdu...tcm8-86022.pdf

Is this what you mean by consequence modelling?
If so, then I think CFD and consequence modelling are totally different aspect of modelling. CFD is a physical fluid dynamic modelling technique, whereas consequence modelling is an estimation of what will happen in the environment after an event starts.

To put it simple, in my understanding and opinion, consequence modelling is a statistical modelling regardless of what is in the 3D physical world, while CFD is a real 3D world physical modelling.

[sbaffini]

Well, in this case, it seems to me that consequence modeling is just a probabilistic framework which is, indeed, independent from the tools used to investigate every specific scenario.

If this is the case, I think that congestion (with reference to pipes, I guess) is only one of several possible scenarios where CFD might be of help, the overall set being, of course, dependent from the specific case.

For example, for off shore platforms, helideck safety is just another example where, depending from the available budget, CFD might be of help. I guess that for oil and gas in general and petrochemical in particular, you might have contaminant dispersion scenarios (where direct CFD is the alternative to the gaussian plume approach and similar ones), an infinitely long list of fire safety scenarios, explosion scenarios, etc. etc.

Congestion is just one possible scenario then, but the list can get as long as your available budget, and possibly longer, as properly done CFD is your most accurate tool to describe any scenario involving fluids.

This is in principle. In practice, your specific industry must have some standards which dictate which scenario require CFD and which ones instead don't. Maybe all your pipes are in the middle of some desert; in that case contaminant dispersion with CFD might just be overkill. Or, for pipes in particular, 1D CFD codes might be better suited than full 3D ones (which might be especially expensive and unsuited for certain scenarios on long pipes).

The FLACS site, indeed, can be used as a guideline on which scenarios may actually benefit from CFD.

But the specific information on what can be considered congested and what not is something which is slightly beyond CFD, or at least it involves the structural side (which might be the object of computation as well) and for how much congestion that was designed. Also, I hope, there might be safety measures in place to mitigate some scenarios while keep the pipe functioning. In this case, the congestion boundary might be blurred and/or depend from additional factors. As an hyperbole, consider that from the pure CFD perspective no pipe is actually congested as long as there is some passage in the pipe... but of course the resulting pressure might be infinitely large.

In practice, all of these aspects escape the basic CFD and require someone who either knows it already or is paid to get into it.