Typical days with storage in PyPSA

[Eva]

Dear PyPSA community,

 

I would like to use temporal aggregation to reduce computational time. But I have not found how to use typical days (with tsam tool for example) while considering storage (with Kotzur et al.’s method for example) in PyPSA.

I would like to use temporal aggregation with MGA method (near-optimal feasible space), and I thought PyPSA could be a useful tool for me.

Can someone help me to know if temporal aggregation can be done in PyPSA and how I could approach it?

 

Many thanks in advance,

Eva

[lisaze...@gmail.com]

Hi Eva,

I have started to implement this into PyPSA a while ago. I didn't follow it up because for our models the additional constraints for the storage were increasing a lot the memory requirement.
You can find my developments in the following branch (already 2 years ago)

https://github.com/lisazeyen/PyPSA/tree/time_aggregation_store

You have to double check the formulation if you want to optimise without pyomo, I think there was still a bug. I know that in the eTrago project they are also using the Kotzur formulation. You can find their code for the temporal aggregation here

https://github.com/openego/eTraGo/blob/dev/etrago/cluster/snapshot.py

Best wishes,
Lisa

[Eva]

Thank you very much for your help, Lisa.