Nice example!

Just a couple of minor comments.

One other thing that this example could include is running a coarser full horizon run (12hrly?), setting the water value based on the result of that and then running rolling horizon at a higher resolution. This would show the benefit when trying to mitigate model distortions when doing temporal simplification (to reduce computational burden) since the rolling horizon is computationally easier to solve.

Similar to #1318, the model created by the example is too large

GurobiError: Model too large for size-limited license; visit https://gurobi.com/unrestricted for more information

We can either rewrite the example to get a smaller model, or add functionality to mark notebooks to not be run in the CI. But there is no guarantee that those stay up to date then

Really nice example @fneum! We are currently implementing the rolling horizon in the open TYNDP, so that's helpful. Maybe it makes sense to rename the jupyter-notebook? Since it does not really show the implementation of hydro storage to something like rolling-horizon-seasonal-storage?

For storage units the right shadow prices from which one would assign the marginal costs is n.storage_units_t.mu_energy_balance, is this correct? If so I can add this information in a text box to your example.