Modelling multiple flow splits within a flow split….

[Dana K]

Is HEC-RAS (steady state) able to adequately model a split flow (flow optimization) with 2 additional splits further downstream? The flow splits will all merge at the same location downstream.

For example, the flow is split with the majority of the flow (Q1) conveyed through the main channel and some flow (Q2) is conveyed through the low lying area (i.e. floodplain), which consists of a well established residential area.

Further downstream, the channel capacity is reduced and the flow in the channel is split again. The majority of Q1 remains in the channel (Q1a) while the remaining flow (Q1b) spills into the floodplain/residential area and joins Q2 (+ Q1b = Q2b).

Another split occurs again further downstream. Flow from the channel is split. Majority of the flow remains in the channel (Q1ai) and the remaining flow (Q1aii) spills into the floodplain and joins the other flow through the floodplain Q2b (+Q1aii = Q2bii).

The flow from the floodplain (Q2bii) merges back with the flow in the channel (Q1ai)further downstream.

The way I have modeled this scenario is:
I) split the flow at the initial split location by utilizing the split flow optimization option. The initial capacity in to the channel was set as the capacity of the channel.

II) For the second split, I just changed/reduced the flow to the capacity of the channel and increased the flow at the associated cross-section in the floodplain.

II) For the last split, I also reduced the flow in the channel to the capacity of the channel and increased the flow in at the associated cross-section in the floodplain.

I’d appreciate any suggestions on how to properly model this
scenario and if my modeling approach is “correct”/representative.
Thanks,
Dana

[Lonnie A]

Dana,
If you can accurately define the reaches where the flow leaves the channel and moves downstream you should be able to model it with the optimization. What you described sounds like unsteady would be the tool I’d use and perhaps model the splits using the 5.0 version 2D flow option. If the flow spills out and covers a wide area or travels a significant distance the flow rate returning could be significantly less than what left.

I have seen models compute a peak flow diversion into a overflow area and when you mapped the overflow inundation area the volume required to fill the mapped area was greater than the total volume of runoff. This is where the use of unsteady or detailed storage routings are a must. Anytime you want a confident answer over areas receiving a diverted flow the consideration of volume is a must.
Lonnie

[[email protected]]

Hi,

I was trying to optimize the lateral structure (Levee) flow and Every time when I optimize the flow, it gives a message of “Split flow optimization failed to converge”. How can I solve it? Also, is there any way to increase the iteration so that the split flow will optimize successfully?

Thank you

Regards,
Indira

[Author]

Posts