It may be helpful to know what you are trying to optimize. For example, the sizing of the stormwater network for a given flow condition, or something more like the routing of flow while the model is running?
Not sure which are really applicable to your system, but some other potential causes I've seen for this include:
[li]External Water Levels set on outlets without the gradual startup enabled[/li]
[li]Network Loads starting up suddenly[/li]
[li]ADP used for filling[/li]
[li]any control elements moving suddenly at startup (e.g., a valve opening fast)[/li]
Basically, anything that fills the model suddenly. If it's happening a few seconds into the model run, sometimes I'll set a very small reporting timestep to observe the instability in a little more detail to try to identify the source.
If it's not related to filling, some other potential causes are that the ground level is set too low or if the node is intended to be pressurized, the cover type may need to be sealed.
Not sure exactly what the large task is you're trying to do, but here's some workarounds I go to when Mike Urban crashes while trying to do something with a large number of elements:
[li]If it's something just involving the tables, try using Microsoft Access, possibly using an Action Query of some type[/li]
[li]If it's something geospatial, use ArcGIS Desktop, it's a little more robust with large numbers of elements[/li]
[li]Compact and repair your database in Microsoft Access before attempting the task[/li]
[li]Close and re-open Mike Urban so the Mike Urban instance is fresh (at least in my experience, the Mike Urban process usually hangs when you exit, so go into Windows task manager, and kill it to truly restart from scratch)[/li]
[li]You may also be able to use the Import/Export tool's Advanced functionality to write some scripting to do what you want for certain tasks[/li]
[li]If you're ok with SQL, you may be able to use the Tools > General SQL Command... tool to directly edit the database (usually, Microsoft Access works a little easier if you're not too familiar with SQL)[/li]
Welcome to Mike Urban :)
Regarding the error message, you can try opening up the table mw_MDemand in something like Microsoft Access and look for nulls/blanks in the values in the two columns JunctionID and Sqn:
[li]JunctionID should be the MUID for the nodes to which the demand is assigned. If there's a blank in this column, try to figure out what node the demand should be assigned to and enter its MUID. [/li]
[li]Sqn is an integer, if there are blanks here and you have more than one demand for a node, just sequentially number them (e.g., 1, 2, 3, 4, etc.). If there are blanks here and you only have one demand for each node, just enter 1 so that it's not blank anymore.[/li]
If everything is empty in a row, it may have been added by mistake and can just be deleted.
I'm not really sure on the Scenario Manager part, it sounds like you're setting it up right. I have seen in MOUSE at least when it throws that error about the "unpredictable behavior" it gives up and doesn't actually change the scenario.
No, I don't believe so. In my experience only one instance of the Mike Urban application will open and it can only load one model at a time. (Though they have a Close All and a Close toolbar button for some reason...this has made me wonder if there's some trick to getting a couple simultaneous models open).
I sometimes use multiple instances of Microsoft Access to edit the Mike Urban model if I need to move stuff in between models and you don't need the geospatial component. You could also work in ArcGIS Desktop, which allows multiple instances. Another Mike Urban tool for working with multiple models depending on what exactly you're wanting to do is the Import/Export Tool with Backup/Restore Simple XML to move model elements between multiple models.
You can have multiple model simulations running, however. And you can edit separate model while a different one is running.
I haven't found a way to select two consecutive pipe elements and have them join with the click of a button, which is what I'm guessing you're looking for. If there's a couple I usually manually merge them, changing the TONODE in one pipe element and deleting the other along with the intermediate node, then fixing all the parameters affected.
If you're doing this repeatedly, you may want to look into Mike Urban's Tools > Simplification... tool, which will merge pipelines for you based on criteria. The user interface is a little clunky (lots of steps), but it can do the actual merging, preserving the pipe element vertices. There isn't really a way to preserve MUIDs across this, so I find myself using ArcGIS Desktop's spatial join comprising a part of a line segment to get the old MUIDs back.
Anyone have a good SQL snippet for finding duplicate/parallel links in the Select by Attribute dialog? I came up with this, but it does seem to include some false positives. Thought I'd check if anyone had a snippet already built before going into a bunch of time debugging why.
[code][MUID] IN (SELECT SS1_mV1.[MUID] FROM msm_Valve SS1_mV1 INNER JOIN msm_Link SS1_mL1 ON SS1_mV1.FROMNODE = SS1_mL1.FROMNODE AND SS1_mV1.TONODE = SS1_mL1.TONODE)OR [MUID] IN (SELECT SS2_mV1.[MUID] FROM msm_Valve SS2_mV1 INNER JOIN msm_Valve SS2_mV2 ON SS2_mV1.FROMNODE =SS2_mV2.FROMNODE AND SS2_mV1.TONODE = SS2_mV2.TONODE AND SS2_mV1.MUID <> SS2_mV2.MUID) OR [MUID] IN (SELECT SS3_mL1.[MUID] FROM msm_Link SS3_mL1 INNER JOIN msm_Link SS3_mL2 ON SS3_mL1.FROMNODE =SS3_mL2.FROMNODE AND SS3_mL1.TONODE = SS3_mL2.TONODE AND SS3_mL1.MUID <> SS3_mL2.MUID) OR [MUID] IN (SELECT SS4_mL1.[MUID] FROM msm_Valve SS4_mV1 INNER JOIN msm_Link SS4_mL1 ON SS4_mV1.FROMNODE = SS4_mL1.FROMNODE AND SS4_mV1.TONODE = SS4_mL1.TONODE) [/code]
I'd take a look at ESRI's Batch Project tool (in the Projections and Transformations toolbox)
I've also conducted this in Microsoft Access with action queries or find/replace, but the nice part about using the built-in field calculator is that some of the related tables will also change the MUIDs (not all of them).