pypsa / nomopyomo Goto Github PK

I calculated the lopf of a small example with 2 nodes and 2 snapshots (attached) and got a different optimal solution. With

network.lopf(solver_name="cbc",formulation="angles")

I get 879719.89 and with

nomopyomo.network_lopf(network, solver_name="cbc")

I get 904301.23. The same is true solving with gurobi. The problem seems to be the computation of the state of charge (in this case of cyclic and non extendable storage units). If I add the line

network.mremove("StorageUnit",network.storage_units.index)

the optimal solution is the same for both methods (904301.23).

Attachment:
nomopyomo_bug.zip

This is not actually a bug. It is jus a question to understand the idea behind nomopyomo.

What are the current limitation of nomopyomo if each time when I run it I receive the following message:

WARNING:nomopyomo:Committable generators are currently not supported.

Does the optimizer still consider constrains like ramps, start up/down cost and so on, or just optimize around the marginal cost?

Thanks for your help.