This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Force main with SewerGems using SWMM engine

Hi,

In my sewerGems model I have a pump station with a force main of 2km and 1050 mm diameter. This force main is discharging in a chamber that is also the start of an inverted siphon (to cross a river).  I am using swmm engine. In the scenario summary, there is a small continuity and routing error (±1%) but there is a high percentage of non-converging (40%). I'm trying to find the reason of this non-converging high percentage. Here are my questions:

  1. I tried to model the force main as a pressure pipe with 6 pressure junctions (because there is slope change in the pipe) and one small virtual pipe between the pump and the first pressure junction. Is it the best way do to it?
  2. For the inverted siphon I used a tank on one side of the river and a transition node for the lowest point and the discharge is a manhole. I use gravity pipe. The results seem ok, but is it the best way to represent an inverted siphon?

Thank you

  • Marie-Joelle,

    What you've done to model with the force main and the inverted siphon sounds similar to what I would have done and it seems like the good way to represent what you have in the system. You might want to take a look at this wiki, however, which suggests how to construct an inverted siphon. There might be some information that could help you. What you can do if you have any additional ideas for configuring these areas is to run a sensitivity analysis. That will allow you to test multiple scenario layouts to see if one is better than the other, but with an overall +- 1 % continuity error this isn't a something you absolutely must do. You can read more about how to perform a sensitivity analysis from this wiki link

    Regards,

    Mark

    Mark

  • Marie-Joelle,

    To add to Mark's answer:

    Regarding the high percentage non-converging: fixing up any data entry or model layout concerns would certainly be a good first step. Once you're comfortable with that, check out the below article on adjusting some of the advanced calculation options like the Routing Step, to achieve better convergence:

    Troubleshooting unstable SewerGEMS and CivilStorm model results using the Explicit SWMM Solver


    Regarding item #1 on your list - typically the Explicit solver works best of the pipe immediately upstream and downstream of the pump is set as a virtual pipe. If you haven't done so already, try changing the pipe upstream of the pump to a virtual pipe.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.