Increasing EG-line

[Caroline]

Hi,
I have a problem with an increasing energy grade line in a model I have taken over. The geometry file have been used for steady state before and works well.

However, dynamic simulation is not working that well. The increase in EG appears over a converging junction. After the junction I have two reaches, A and B. In B there is a high weir and in A a lower, resulting in larger flow in A than B.

Where B starts, the EG line stays almost constant over the junction, but where A starts the energy grade line jumps up. The jump increases as I lower the weir.

The EG-line sinks marginally if I do increase Manning’s number in reach A.

I have already interpolated extra cross sections both upstream and downstream the junction with no effect. This junction uses the energy balance method. I also looked into Htab.

I’d like to get some suggestions for how to solve the problem.

Cheers,
Caroline

[Anonymous]

Regarding the downstream increase in the EGL that HEC-RAS is reporting: This often happens in association with a non-convergence of the unsteady flow solver or a critical depth solution where the program makes an assumption about the solution and continues. That assumption overrides the governing equations. These situations can result in a solution that shows the EGL not decreasing in the downstream direction. However, the requirement of a downstream decreasing EGL is based on steady flow hydraulics for which the energy conservation equations are used. The unsteady flow equations are based on momentum conservation. Note that, when doing an unsteady flow simulation, RAS reports the EGL from a snapshot in time using the steady flow definition of the EGL at each cross section. Usually that EGL can be used to evaluate how good the solution is working; but it’s not a complete description of the unsteady flow solution because it does not contain the acceleration terms in the momentum equation. These might be subtleties that may not be what you are seeing. To further diagnose what’s going on in your model, suggest that you turn on the “Computational Level Output” and under the Output Options select “Differential Equation Parts”; this output will give you the ability to look at the magnitude of the various terms in the unsteady flow equations. This info. might help diagnose why the EGL is showing the behavior that you see.

[Caroline]

Thank you! Your answer helped a lot for understanding.
I have started to look into those equations.

But first, I did some changes in the geometry file. When removing ineffective flow areas immediately upstream of the junction I got rid of the most increase. The rest could be removed by setting the exact same values of Manning’s number in the reaches on both side of the junction. The latter is unfortunately not consistent with the real geometry, which have walls after the junction and a river bed before.

[Author]

Posts