Modeling drainage during a deluge event

[CKauf]

Hello everyone,

This is my first post, but I have been stalking the forums for a while and generally find very helpful information here.

So I have a problem, the company I work for has internal standards for our secondary containment. Our insurance provider require this type of strict standard, so I can't just ignore it. Currently, in one of the process towers I work with, we have more deluge water than our secondary containment can handle. The problem is, the distribution of sprinklers is not uniform; vessel protection requirements make use have more sprinklers around the larger vessels. So my project team has devised a plan to divert half of the deluge volume from one spill pond to another spill pond. While keeping the other half of the flow in tact.
I am looking for suggestions to model where the water will accumulate (and ultimately drain), so I can adequately defend the project when it comes up for review. The primary issue is due to the variation in sprinkler pattern and the obstructions in the tower. I need a software that can generate a 3d model that can simulate this and then can help check and make sure the new drainage system will be adequate. It also needs to be able to model structural steel.

If anyone knows of a software that can do this, please let me know.

TLR - Need a program to model fire water flow, accumulation, and eventually drainage. Needs 3D capability due to obstructions in the building and location of where the sprinklers are (not uniform flow due to three different types of sprinklers in the building).

Thank you for your time.
Sincerely,
CK

 

[CKauf]

One other thing to note, software suite doesn't matter.
My company utilizes both Bentley and Autodesk.

 

[cvg]

not a perfect match, but HMS could be used and it is free. I'm thinking 3d modeling might be overthinking things a bit.

http://www.hec.usace.army.mil/software/hec-hms/features.aspx

 

[gbam]

2-D model like FLO-2D could work. There is a free/basic version that can evaluate the dispersion of sprinkler water if modeled as precipitation, maybe? you will need to code in the topography as a grid and then develop a precipitation model that matches your sprinklers.

http://www.flo-2d.com/

Good Luck & Let us know how it turns out.

 

[CKauf]

cvg, I took a look at HMS and was unsure that I can make custom precipitation patterns. If you know a little bit more about the program, and can correct my judgement, please do.

gbam, I have a question about flo-2d, the reason I would want to go 3d is because of the vessels in the building causing flow disruptions and creating voids in the pattern. Do you think that just leaving their footprint on each floor as a void would be effective? Or would I need to find something else to model it with?

 

[cvg]

HMS can model any precip pattern you want
sure, flo-2d could work but your grid would need to be very small. coding that in would be time consuming and then there is the learning curve which generally requires tech support (which is not free)

 

[gbam]

What cvg says about the learning curve is true, oh so true! In FLO-2D you can model sinks as nodes to represent a drain, per se, you can model tanks as obstructions or give them higher elevations. The detail in which you model will adversely affect the computation time; ie. more detail more computation time. Not sure if it can model the differential precipitation as described in your sprinkler note. As to the obstructions the 2D flow will disperse the flow around the obstructions following the topography of the ground/floor.

Might I suggest you step back and ask if this amount of detail is necessary. Maybe if you post a layout schematic we can see the big picture a bit better.

 

[racookpe1978]

Have you looked for any "mold pouring" or mold filling programs?

Custom side: Make an accurate 3D model of the tanks and the perimeter holding walls. Make it as accurate as reasonable, then model your rainfall as cubic inches (cubic feet or meters or cm^3 as you see fit) per hour.

Multiply by the area of the perimeter walls (we are assuming there is no overflow from any adjacent tanks, right?) that gives you a worst expected case "fill rate".

Then model that fillrate (volume/hr) back into a inches-per-hour rate into the separate perimeter sumps as a "volume" in the bottom. If your perimeter dams are wedged shapes (not concrete vertical walls) the fill rate will be slightly lower for each hour since the area to be filled increases as the wedge slopes.

Then, after a number of hours, the primary fills up. That flows into the secondary pond, but the secondary pond has at - at a minimum, its area x nbr of hours to fill the primary already loaded. Subtract the difference = number of hours left before the secondary flows over.

 

[CKauf]

racook,

I am not sure what you are talking about?
I know the flow rate of the deluge systems in this building. I am just trying to make an intelligent model to simulate an event and figure out the volume of deluge each square unit receives per minute.
I will then take this information and use it change the drainage in the building so half the water goes to one drain and half goes to the other.
I've already tried splitting this up by systems (currently 4, soon to be 5), the problem is there is significant overlap between the systems.
Also, secondary containment is when chemicals are released (primary containment is the intended vessel), this is the definition that NAFPA uses.

I really appreciate all the help everyone, have a nice night.

 

[CKauf]

racook,

So I had some free time this morning and started looking into a mold filling program to figure this out. I am unsure as to which programs are good and which ones aren't, any input would be appreciated.

 

[beej67]

I'm confused why this needs to be that complicated.

If you're modeling sprinklers and not rainfall, and you know the flow rate of the sprinklers, then why bother with a runoff model at all? You already know the flow rate, because you control the sprinkler, yes?

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[CKauf]

beej67,

You are absolutely correct, we do know the flow rate of the sprinklers, but the systems are heavily overlapped. With this project the issue we are running into is that we need to be able to quantitatively prove that after this project we are redirecting 50% of the volume to a different set of trenches and ultimately spill ponds.

 

[beej67]

I would think you could do that by drawing an individual 'watershed map' for each sprinkler system, identifying which percentage of each system makes it to each pond. Multiply those percentages by the flow in each system (which you control) and you know the flow into each pond from each system. Add them up.

Is there a reason this simple arithmetical approach won't work for your purposes? It's a brute force approach for sure, but it'd still probably take you less time than trying to even enter the data into something as complicated as Flow2d.

I'll admit to having little experience with your particular problem, so if I'm asking dumb questions please forgive me.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com