PIPENET transient model result is different with Spray/Sprinkler model

[eriskawk]

Hi everyone,

I am currently doing modelling of a firewater system at LNG Terminal & Regas Facility using PIPENET both Spray/Sprinkler module (steady state) and Transient module (dynamic) for surge analysis. This is my first time conducting surge analysis using transient module.


In the Spray/Sprinkler module, I got pump discharge pressure at the most remote area = 11.8 bar.
Then I extracted this module into Transient one to conduct surge analysis, and modified some item by only adding non-return valves and deluge valves.

At first I ran the transient module without adding any closure, to ensure that the model is valid before adding more variable further. In this case, I got the discharge pump pressure = 14.96 bar, which was way bigger compared to what I got in the spray/sprinkler module (11.8 bar)

My question is, isn't the result supposed to be equal? As there was no closure added, means it was actually running on steady state condition. (The graphic result is also flat)

Or how is it supposed to be?

Any response is highly appreciated.
Thank you!

Cheers,

 

[1503-44]

What is PipeNet's definition of closure? Normally in transient analysis closure refers to the degree of accuracy you allow the next timestep solution to vary from a more precise solution. For example all pressures might be deemed accurate enough if the solution obtained between time steps of 1 second are less than 10 psi, 5 psi, or 1 psi. Selecting higher allowed variation allows greater time steps to be used in obtaining the solution to the next timestep, but at the expense of accuracy.
No closure setting may default to some maximum timestep, which may not give you highly reliable solutions.

I do not use pipenet, so that's just only a WAG at what you might be running into. If you decrease the allowable pressure variance "solution closure", does the solution more closely align with your expectations?

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[MortenA]

@eric,

Usually a transient models starts from the SS solution, so yes if the trends are "flat" but the pressure differs by this much something looks fishy. I never used pipenet - but other software and sometimes i would find some weird differences in reporting e.g. that the SS would be relative to grade by default - but the transient would be relative to "zero grade" (i remember that this cause me a lot of pain similar to your so thats why i point it out now)

Best regards, Morten

--- Best regards, Morten Andersen