Solver Error: "ReadOpenChannelLinksFromFile - ModuleRunOff - ERR91 - How to solve?

Hello,

When performing a 2D simulation in the SewerGEMS software (10.04.00.158) I got the errors in the image below, including the error described in the title. I attach my used file.

File

Parents Reply Children
  • Hello Luis, bom dia!

    Well, good news for all of us. Today I turned on my laptop and went to do the simulation to save the package and, for some reason that idk, it worked. Now, I would like to know how to improve the visualization like in the image you provided at the beginning of this post. Below is the link with the package and an image of how it looks after the simulation (nothing very visually noticeable).

    package: drive.google.com/.../1kdrD1CCEb7ymnAqU2RXW3YepamqTdoni

  • I can see that your cross-section labels still have "ç" and "~". The error that was happening was again because of this. From the model you've sent, it's strange that it worked. I can see that the connections between the open-channel and the 2D grid were not created, but I cannot reproduce that myself. We will have to wait for you to get another updated version for us to get more information. I realize now that you only got the latest cumulative patch for SewerGEMS 10.04.00.158, not the version that can help us get more information. 

    Can you relabel all elements that have Portuguese characters, accents, etc. and try again. 

    Kind regards,

    Luis

  • I ended up creating copies of the same file to do some tests and solve the problem. It may be that I made this change to one of the copies that I deleted. Anyway, I changed the section naming again and everything appears to be working. I was thinking that maybe the solution to this was the update that Jesse gave me earlier. When I installed it, I don't remember restarting the computer (or the program) before computing again, which is why the error may have still occurred. I have already updated the package inside the drive.

    One thing is that in fact, I noticed that the catch basins are not receiving water input from precipitation, and water seems to be only being indicated at some points in the system. Should I adjust the height of the cath basins or do I need to adjust the topography file in some way (if so, how would I do that?)? I noticed that some profiles have shown some very strange depressions that are worrying me a lot, as seen in the images below.

    (in channels i know that i really need to make some changes)

  • Hello Luis,

    The lack of flow into the catch basins is likely related to the grid elevations in nearby cells compared to the rim elevation of the catch basins. You can see some of this in the screenshot of your profiles above, but using the Grid Browser to review the elevation in different cells will also show this. 

    You set the catch basin and manhole property 2D Surface Elevation Matching to "Adjust Grid to Rim." This is effectively increasing the elevation off one of the grid cells to a higher elevation than surrounding cells. Unless the water elevation reaches a higher value (in some cases as much as 6 meters), surface flow will not be able to reach the rim of the catch basins. 

    This may be helped, in part, by adjusting the grid cell size. This is currently set to 10 meters, which may be relatively coarse compared to your terrain model. As an example, a grid cell size of 10 meters may be larger than the width of a street as represented in the terrain model, which could mean that the flow to the road could be missed entirely. Any higher elevations for building data may also be less accurate too. Try using a smaller grid cell size to see if that helps with some of the results. This will increase the model run time.

    If adjusting the grid size doesn't help, you will want to review the rim elevations compared to the terrain model to try to make this as accurate as possible. 

    The current grid also has a uniform Manning's n value. If the terrain model Includes changes in elevation that represent roads and buildings, you may want to consider including a Surface Definition shapefile to account for changes in Manning's n for these areas. The 2D Modeling wiki for SewerGEMS includes general information on these. 

    Finally, the terrain model does appear to small areas with depressions in it. You can see this in the Grid Browser looking at maximum water elevation. If this is not accurate, you may want to review the terrain model data for these areas. 

    Regards,

    Scott

  • Ok. I believe I can visualize it better now. My impression is that I need to adjust these points (in yellow) to a height that is compatible with the drainage elements. I selected the "Adjust Grid to Rim" option (I don't remember where) precisely so that there would be no cases of falling below the maximum quota of these elements. Wouldn't there be a way to correct these points to try to adjust for a more uniform unevenness through point triangulation or something like that?