You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently renewable profiles and load are added to the high-resolution base network, then it is clustered. This makes the clustering highly dependent on the substation layout of the base network.
If we separate the clustering from the network, we are free to cluster on e.g. custom shapes, or cluster based on weather cells instead
Generating renewable profiles after clustering requires fewer computations (num clusters instead of 5000 base substations)
It would be easier to switch to allowing multiple VRE LCOE-classes inside each region rather than tying them to sub-regions.
We would avoid the weirdnesses that some assets are added before, some added after the clustering - we could uniformise around PyPSA-Eur-Sec-style "assets added after clustering". This would simplify the workflow considerably.
Downside: if you want to cluster based on VRE time series, you have to do extra work (but at least you're not tied to substation layout - you could e.g. cluster time series from the original weather cells)
The text was updated successfully, but these errors were encountered:
The text was updated successfully, but these errors were encountered: