Allocation controlled TabulatedRatingCurve

[verheem]

What && Why
Multiple users have reported they want to use allocation with TBR's, where allocation decides if a TBR may flow or not. It makes sense they expect it already to work since TBR accepts a control state. Often they try to give the control state "Ribasim.allocation", but this has no effect.

How
Allocation should be able to turn a TBR on or off.

[verheem]

Figure out a way to do this without binary variables

[gijsber]

if that is not feasible, we probably have to make life easier to users by offering some handy other method in e.g. ribasim_python

[Fati-Mon]

@AKhakurel3

[verheem]

After a fruitfull discussion with @visr we have come to the conclusion that it makes most sense if we use the Q(h) relation of a TBR as upper bound for the flow when the control state "Ribasim.allocation" is configured.

If there is a water shortage in the basin upstream of the TBR for example, due to a min level demand, allocation will set the flow Q_alloc <= Q(h) such that the level demand is respected. It should be noted that if a downstream basin has a min_level that is not met, allocation will not push more water through the TBR than the Q(h) relation allows.

It should also be noted that the resulting flow of the TBR can them be anything between 0 and Q(h).

[visr]

Yeah I quite like this. My hydrological interpretation is that the Q(h) of "Ribasim.allocation" would be the "open" setting of the gate or weir. You cannot control it to get a higher Q. You can close it as much as you want to reduce the Q as much as needed. The only unrealistic part of this is that you can fully close it, even if h goes to infinity. In reality, at some point the water will flow over the top, even if you'd want to stop the flow for water users upstream. But perhaps it is sufficient to have a parallel TabulatedRatingCurve with a Q(h) that represents unstoppable flow over a closed weir or a spillway.