Analysis of complex piping

[Sloan Bayle]

Hello everyone,

How do engineers design complicated piping systems, such as systems with many branches and equipment? Is there some software that helps calculate system curves for complicated piping? I don't see it being feasible to perform calculations by hand, even with Excel or Matlab. Do people really even draw system curves? I could see how estimates can be done with fair accuracy, but I don't see that being ideal at all.

Thanks for any answers!

 

[MJCronin]

What you seem to be describing is piping network analysis ....

Software for performing such analysis has been around for decades

http://www.aft.com/products/fathom

MJCronin
Sr. Process Engineer

 

[zdas04]

There are pages of adds for various network pipeline models in all of the pipeline magazines. I've never used the one that MJCronin linked, but I have used dozens of others. For the most part what you can say about all of them is:
[ul]
[li]There are people who prefer it[/li]
[li]There are people who hate it[/li]
[li]The software has bugs in it (even something that has been around for decades like Stoner)[/li]
[li]Some packages have strengths that correspond with your needs[/li]
[li]Some packages have weaknesses that correspond with your needs[/li]
[/ul]

At the end of the day, you need to determine what is most important to you (simple looped lines, ease of inputting a compression or pump station, complex equipment arrangements that can reverse the direction of flow in a line, variety of available flow correlations, ease of input, graphical vs. tabular output, etc).

I always take a set of data that has the characteristics that concern me and try to replicate the field results in the network model. The data set is tough, and most fail (nearly all tend to fail in a flow-reversal scenario, but that scenario is reasonably uncommon). A model that matches my data set can (and probably does) have other bugs that I didn't test (e.g., I've never had a network that I was evaluating that had pressure reducing valves, I just don't work on that end of the value chain so I've never tested that feature in a model), but I call it "good enough" and use it to evaluate piping condition for a few systems before I will rely on it for a network design analysis.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[Sloan Bayle]

Thank you both for your reply. I am a newly graduated M.E. and entered a process engineering position. I’m working on a small project now (can be done fairly easily by hand), just was wondering how more complicated projects get designed. I am interested in learning more about these types of software, but won’t ever need for my position.

Thanks again for your help.

 

[PipesandPumps]

There are a few good options out there. The AFT software mentioned above has been around for a while and works. I prefer one called PIPE-FLO myself because it handles control valves much better and has the ability to use actual control valve data and control schemes. If your project is small you may be able to get away with their free version. It allows you to build small models and calculate them. I added the download link below.

https://eng-software.com/download/business-development-demo

 

[swAFT]

Our control valves are set to a specific value and maintain that setpoint unless they fail. Then they will fail in a realistic manner based on the valve type. We have not heard that people want anything else. If you would like to learn more about AFT Fathom capabilities to model control valves, feel free to contact us at info@aft.com