Run Calliope multiple times with historic data before generating results
hiddegrootes opened this issue · comments
What can be improved?
Hi,
I was wondering if it is possible to run Calliope multiple times with historic data before generating results. For example use weather data for PV or wind generation from 2000 until 2023 (as you cannot average hourly weather data on specific data's in different years) in this way the averaging happens after running and the results should be more robust. Let me know!
Version
0.7.0-dev
Hello @hiddegrootes!
It is definitely possible, but it might be something that you have to implement on your side within your workflow.
Calliope is meant to be a generic framework, so separation between specific studies and the general setup is quite important. We could introduce this as an additional feature, but as far as I am aware it is not in the current planning.
A relatively easy method would be to run for each year in sequence, and freeze the minimum installed capacity to the given result after each run. This will net you a very robust system (see https://www.sciencedirect.com/science/article/pii/S0360319920348606 for an example with a different model).
However, keep in mind that this type of robustness tends to lead to very expensive systems, so your mileage may vary depending on the case.