Unsteady Flow model with negative flows and velocity..HELP

[cagood]

Hey everyone,

I am new to the forum. I am a Water Resource Engineer in Denver, CO and need some guidance on a HEC-RAS model I am working on.

I am modeling a "sunny-day" dam breach and have a hydrograph (generated in HEC-RAS)for the dam breach which I am attempting to route downstream via unsteady-flow. The dam breaches and is routed downstream about 30 miles, at which point there is a confluence with another river, 20 miles downstream of there, my reach converges with another large river, and finally 30 or so miles downstream of there, a third and final confluence is seen.

So I am simply trying to route my dam-breach hydrograph downstream while adding in the lateral inflows at their respective river stations.

I am running the unsteady file for 5 days, with a small time-step. A previous model using the same reach was modeled with a much larger dam-breach hydrograph, with much larger lateral inflows from the confluences, and that model produces results that seem acceptable. My current model with the smaller hydrograph and inflows, will not run correctly. I am getting huge fluctuations in both flow and velocity, with times of negative flows and velocity.

Does anyone have experience with this? What am I missing? This model will run fine as well, if I dramatically increase my lateral inflows while keeping the smaller hydrograph.

Any help is appreciated!!

 

[beej67]

Guess:
Is your negative flow/velocity at the beginning and the end of the run, near the confluences of the lateral inflows?

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[chicopee]

What would negative flows and velocities suggest?

 

[beej67]

^^They suggest the tailwater is higher than the headwater, and therefore flow is going backwards, at least in my world. I honestly can't recall if HEC-RAS is a one-way model or not, though. Francesca would know. So would Peter Smart. So would the HEC-RAS documentation.

There might even be a toggle somewhere in HEC-RAS to prevent backflow, and flipping that toggle might fix the model somewhat, although it wouldn't fix the underlying problem causing the backflow. I know they have backflow prevention options in XP-SWMM, so they probably have them in EPA-SWMM.

Describe these "huge fluctuations" cagood. Does your hydrograph look like an earthquake just hit on a seismograph? If so, your model is unstable.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[cagood]

Problem solved. My cross-sections were too far apart. The water surface was crashing in between the large distances between sections.

Thanks for the replies,