Bart de Koning
Bart de Koning
These refactoring points came out of the knowledge sharing session (which was more of a code review session): - [x] Use function `is_main_network` here: https://github.com/Deltares/Ribasim/blob/61b3a173155647b78735a73357908cf8c48ed9ce/core/src/allocation_optim.jl#L1035 - [ ] Introduce a...
## The objective function @jarsarasty suggested the following linear objective function that also achieves equal ratio allocation. Say we have demands $d_i$, allocated flows $F_i$, and the allocated fraction $f$....
Fixes https://github.com/Deltares/Ribasim/issues/1473. To do: - [ ] Update `TrapezoidIntegrandAffect.integrand_value` in `update_basin!` for integrands that contain the basin forcings (both for flow integration and bmi integration) - [ ] `TrapezoidIntegrandAffect.integrand_value_prev` can...
Allocation should not send flow trough model nodes that are inactive (`node.active = false`). This also means either that these nodes have to be represented explicitly in the allocation graph,...
Functions from [this package](https://docs.sciml.ai/FindFirstFunctions/dev/) are internally used in `DataInterpolations`, so if lookups in sorted lists become a performance bottleneck for us we can use them too.
There are a few use cases for testing which are quite expensive: - Run models with various solvers and compare their results; - Run models with a sparse Jacobian (https://github.com/Deltares/Ribasim/pull/459)...
@Fati-Mon requested for one of her models that we support transient `UserDemand` return flow. This is currently not supported, but it is already possible to put it in the `UserDemand...
As explained in https://github.com/Deltares/Ribasim/pull/1444, the approach for integrating flows proposed in https://github.com/Deltares/Ribasim/issues/1431 is not feasible. The next idea was to use [integrating callbacks](https://docs.sciml.ai/DiffEqCallbacks/stable/integrating/) now that these have more support. However,...
This is an idea that came up after observing that https://github.com/Deltares/Ribasim/issues/1431 is very bad for performance, see also https://github.com/Deltares/Ribasim/pull/1444#issuecomment-2107871589. So my thought was: What if set up a parallel ODE...
The plotting for some of our test models that are just a straight horizontal line looks pretty bad, for instance `two_basin`:  Maybe the easiest fix is to compute the...