1D/2D model losing flow / gives low WSE

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

1D/2D model losing flow / gives low WSE

Andy
I have a 1D/2D model which includes a 1D reach downstream of a 2D Flow area with an inline structure just downstream from the 1D/2D connection. There are three constant inflows to the 2D flow area totaling ~ 4,500 cfs.

I have set gates the inline structure to open at WSE 1115' (start closed) with no pilot flow. Model run results show a WSE drop from 1115' (initial condition) to ~1100' - 1107' at the dam / within the 2D flow area. Also, at the 1D/2D connection the flow drops to a large negative flow before shooting back up and leveling out around 900 cfs. The gates never open because the WSE remains in the range described above, not high enough for gates to open.

I am thinking that the issue has something to do with the 1D/2D connection as I was previously able to run the model successfully without the downstream 1D reach. This model is intended to simulate a steady state in preparation for a dam breach plan.

I have:

Set initial conditions for WSE/Flow where applicable at boundary locations, for the 2D flow area and at the 1D/2D connection
Set iterations (up to 20) for 1D/2D boundary computations
Set a "long" Initial Conditions Time (72 hours)
Reduced computational time step
Added additional cross sections just downstream of the 1D/2D connection

Any ideas of what could be going on here / how to fix? The model is "stable" but producing the results described above.

Reply | Threaded
Open this post in threaded view
|

Re: 1D/2D model losing flow / gives low WSE

jarvus
You mention that you have a long Initial Conditions Time.  Have you also tried a large number of Warmup Time Steps?  It sounds like it is unstable at the 1D/2D connection for the very first time step.  A warmup period could help with this.
Reply | Threaded
Open this post in threaded view
|

Re: 1D/2D model losing flow / gives low WSE

Andy
I tried adding Warm-up Time Steps and along with some other tweaking I'm getting much better results - thank you for the suggestion!
Reply | Threaded
Open this post in threaded view
|

Re: 1D/2D model losing flow / gives low WSE

Lonnie A
Did you set a initial stage in the 2D?  Is it may be the volume is filling the 2D initially?  With the warm up you are probably filling the 2D area ahead of the run which is why you are getting better results?