Every simulator on the market (and there are hundreds) has bugs and limitations. Some are major (e.g. in PipePhase if you guess wrong on flow direction it will pretend to run and then give you really wrong answers that require line by line review to find), some are cosmetic.
I think if I was replacing my venerable MNET (which is no longer for sale) I would probably go with Schumberger's PipeSim, but it is a near thing.
Any simulator that you buy should work for your problem. Always load an existing system and calibrate it for a real set of flow rates and pressures. If you can't match every point in the system to within a couple of psi then either your "known" data set is broken (which happens far too often), your description of the system is faulty (again, real common), or the simulator doesn't have the knobs you need to tweak to match your local anomolies (e.g., particular lines full of liquid). If you can't match your toughest system then run away from that program because you'll certainly eventually need to evaluate a modification to your toughest system and if you can't match current conditions how do you think you'll be successful extrapolating a change?
One uncommon thing I did with MNET is re-calibrate the model quarterly with new data to evaluate the pigging schedule. If a line needed more negative adjustment then we weren't pigging it often enough. If it needed a lot less them maybe we were pigging a bit too frequently. If the calibration process had been a lot harder then I'd never been able to do that.
I only listed the last two because he used the word "burner", so I assumed he was not talking about a "city" domestic gas supply network. I think its some kind of fuel system network in a plant. From what limited knowledge I have of those last two programs, I believe they should be able to handle that.
I work with HYSYS and winsim a lot - and they are fine for networks of limited complexity (1-2 branches tops). Bescause they cannot "work backwards". This means that you have to set up a lot of "adjusts" and such in order to get "same pressure" at brances - whereas a "real" flow program such as pipeline studio will solve the whole network.
![[worm]](/data/assets/smilies/worm.gif "[worm] [worm]")
