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

Deferent engine Results in SewerGEMS

I am getting big difference when I run the Storm Network model using Bentley EPA SWMM than the supplied EPA SWMM 5.0.12, it is the same model and the same parameter, I just switch between two engines; 

I got 8.17 m3/s at the outfall by using Bentley EPA SWMM, and I got 10.55 m3/s when I run the model with EPA SSWMM 5.0.12. That is a huge difference. the model shared for troubleshooting (Bentley EPA SWMM & EPA SWMM).

Why that difference? which one is more accurate and to be used? please justify both results.

Bentley EPA SWMM

EPA SWMM:

 .

Regards .... Mohamad Azzam

  

  • But when I run the model by using EPA SWMM, there are flow going to the gutter. Do you mean that flow is the excess gully capacity 100% goes to the next gully thru the gutter?

    Almost all of your gutters have zero flow during the entire simulation, because when using the EPA version of the engine, catchbasin elements capture 100% of the connected surface flow (since as explained previously, EPA-SWMM itself does not have an inlet capture capability). 

    There are a few gutters that do have flow and these are because of catchbasin overflow. When a catchbasin floods and the hydraulic grade rises above the rim elevation, overflow passes into the downstream gutter if it exists. Below is one example. See more here: Modeling catch basin overflow into gutter

    And when I used the enhanced EPA the gully capacity percentage will be considered and that percentage will enter to the system and the rest will go to the next gully by gutter ?

    If you mean the "SWMM (BENTLEY 5.1.13)" option in the "Solver Compatibility" calculation option then yes - we have enhanced the SWMM solver to add the capability to calculate inlet capture whereby a percentage of the flow is captured (and enters the subsurface pipe system) and the rest bypasses the inlet and passes down the next gutter.


    Regards,

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

    Answer Verified By: Mohamad Azzam 

  • So, if a catchment is directed to a "node" in EPA-SWMM, all of the flow from the catchment enters the system. Meaning, 100% of the catchment flow is "captured"

    But when I run the model by using EPA SWMM, there are flow going to the gutter. Do you mean that flow is the excess gully capacity 100% goes to the next gully thru the gutter?

    And when I used the enhanced EPA the gully capacity percentage will be considered and that percentage will enter to the system and the rest will go to the next gully by gutter ?

    is that correct?

    Regards,

    Mohamad

  • Mohamad,

    The EPA-SWMM software does not have inlet elements or inlet capture calculations. So, if a catchment is directed to a "node" in EPA-SWMM, all of the flow from the catchment enters the system. Meaning, 100% of the catchment flow is "captured" - this is why we assume full capture in this case if you are using the EPA-SWMM compatibility engine calculation option (to match what you'd get in EPA-SWMM). You can see this for example if you export the "office project" sample model (which has on-grade inlets) to SWMM5 format and import into EPA-SWMM 5.1. 

    SewerGEMS is able to perform HEC-22 inlet capture calculations and inject the resulting flow into the system (to be solved by the SWMM engine) based on the calculated capture efficiency. Since EPA-SWMM itself cannot do that, SewerGEMS ignores those calculations when setting the engine to EPA-SWMM, so that the results are comparable to EPA-SWMM. I have added some additional clarification to the related Wiki article.

    So - unless you have a need to produce results that are compatible with EPA-SWMM itself, then since your model uses on-grade catalog inlets, you should use the "SewerGEMS" engine version.


    Regards,

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

  • Thank you Jesse.

    Please correct me if I am wrong; you are saying EPA considering the On Grade gully is full capture (100%), while Bentley Enhanced calculates with considering the efficiency that we used in the input? if so; you mean that when we are using EPA there is no difference between "In Sag" and "On Grade" and the EPA deals with both the same; am I correct?

    On another hand, if the EPA considering the full capture for "On Grade" how are the gutters show and convey flow?

    Regards,

    Mohamad

  • Hi Mohamad, I see you uploaded a copy of a model so I took a look. Most of your catchbasins (1593 out of 1732) have the inlet location set to on-grade. You can filter or query on the "inlet location" field to see this. As mentioned earlier, the "EPA" version of the SWMM solver does not support HEC-22 inlet capture calculations and assumes full capture, whereas the "Bentley enhanced" SWMM solver does support on-grade capture calculations, so results will be different due to the difference in efficiency.

    For illustrative purposes if I global edit all catchbasins to be in-sag and re-compute (I set up two scenarios with different calculation options where the solver version was changed), the results are the same.

    Note that it appears you are using a slightly older version of SewerGEMS: 10.02.01.04. For upgrading, see: Downloading OpenFlows Software


    Regards,

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