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

SWMM 5.1 Results differ from Sewergems SWMM Calculation Results

I am getting a -21.0% Flow Continuity Error in my Sewergems model so I exported it and ran it in the SWMM 5.1 software and got a -0.16% Flow Routing Continuity Error.

I'm wondering what could cause such a drastic change in results. 

A few notes. The model has two ponds modeled and one outfall. One inlet that is upstream of the pond has a -800.76% continuity error for it's surface node (in sag inlet - "Node 828f_Surface). Obviously this is causing the most problems but I'm not sure how to fix it since the inlet is not overflowing or surcharged. The rest of the nodes listed in the "Highest Continuity Errors" section within Sewergems are in the 9-36% range. 

I am using Sewergems CONNECT Edition Update 1 (10.01.01.04) within Microstation V8i SS4 and EPA SWMM 5.1.012. I will try to upload the "Outfall D6_NEW" file for reference. 

  • Thanks Scott. I've moved forward with a smaller orifice which does reduce the error quite a bit and actually helps the pond perform better in general. 

    I think I understand what you're saying though, regarding setting a small area of the pond at the elevation of the lower orifice, since a thin "sliver" of volume between the pond invert (bottom of side slopes) and the lower orifice elevation will theoretically exist, but I'm currently ignoring it to be conservative. I believe I could add a row into the pond's area-elevation table for the lower orifice elevation, and even just assume a small area equal to say 10 sf. This would at least allow the model to correlate the "bottom" of the basin with my lowest elevation at the orifice. 

    Thanks again for looking into this and providing feedback. 

  • Hello Jared,

    I have some additional information on this issue from the developers You may want to review the results in EPA SWMM, because even though the continuity error may be small, the results may not be accurate.

    In EPA SWMM, the developers report that the EQT curve when the pond is at an elevation of 25 feet shows a flow of only about 7 cfs. This is at odds with results in SewerGEMS using the Explicit solver. As a test, you may want to try setting the orifice elevation equal to the bottom of the pond and see what the results look like.

    I am not really able to picture what this pond looks like, but if there is pond volume sloping down to the orifice, would it be possible set the elevation of the pond all the way to the orifice elevation, with a small volume up to the original pond invert of 21.5 ft? Maybe this isn't possible, but it might be worth considering.

    Regards,
    Scott

    Answer Verified By: Jared Sandberg 

  • Hi Scott - Thanks for the feedback. 

    Increasing the time step or output increment did not seem to help so I've gone with reducing the orifice to a size of 4" instead of 6" which reducing the error down to about 9%. 

    Regarding the elevations of the orifice, I wondered how the model would handle the situation so perhaps it is contributing to the problem, but what we were trying to do is model a sloped pond with an orifice that is a certain elevation (bottom slope of XX% toward the outlet structure) lower than the "bottom of the pond" per the model. Hence setting the elevation of the orifice and the FG at the outlet structure below the bottom of pond. It does seem to be okay in other models where we are doing the same thing, and I'm not sure how the pond and outlet structure export to SWMM, but again, the error in the SWMM 5.1 program was essentially 0%. 

  • Hello Jared,

    In addition to the information above, the developers noted something else that may impact the stability of the results. It appears that the two ponds in the model have composite outlet structures that include elements where the elevation is outside of the operating range of the ponds. Each of the two ponds have an orifice as part of the composite outlet structure, but the invert elevation of the orifices is below the lowest point of the associated pond. I would review these and perhaps other elements in the model to be sure that the input is reasonable. 

    Regards,

    Scott

  • Hello Jared,

    There were improvements made to the Explicit solver in relation to the the SWMM solver with regards to pond outlets. This is the reason that the operation and results are a little different. The results tend to be better in SewerGEMS for cases like this. 

    There are a couple of things you can try to improve the continuity error in SewerGEMS. If you set the the time step to 5 seconds and the output increment to 120 seconds, there is an improvement to the continuity error. Setting the time step to 10 seconds and the ouput increment to 5 minutes improves it further. 

    Please let us know if you have any questions.

    Regards,

    Scott