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
What
There is already initialization of the optimization problems for subnetworks. This is also needed for the main network, which is slightly different.
How
Make sure that the model for the main allocation network is initialized first (just do them in ascending order by allocation network ID)
When initializing the main allocation model, collect edges to subnetworks (allocation_network_id != 1) in Allocation.main_network_connections.
Each edge to a subnetwork is treated as a user in the objective function and constraints of the optimization problem for the main network. To clarify: yes, this means that subnetwork can be represented by multiple 'users' in the optimization problem for the main network if there are multiple edges connecting the main network to the subnetwork. This is because the subnetwork will have different demands for each such connecting edge determined by the subnetwork structure. The choice of how to treat the distribution of demand from a subnetwork over these multiple connecting edges has a significant effect on the behavior of the solver, I think especially for non-linear objective functions.
In the optimization problems for the subnetworks, add constraints for the supply/capacity coming from the main network.
As discussed with @visr, users within the main network are allowed.
The text was updated successfully, but these errors were encountered:
What
There is already initialization of the optimization problems for subnetworks. This is also needed for the main network, which is slightly different.
How
allocation_network_id != 1
) inAllocation.main_network_connections
.To clarify: yes, this means that subnetwork can be represented by multiple 'users' in the optimization problem for the main network if there are multiple edges connecting the main network to the subnetwork. This is because the subnetwork will have different demands for each such connecting edge determined by the subnetwork structure. The choice of how to treat the distribution of demand from a subnetwork over these multiple connecting edges has a significant effect on the behavior of the solver, I think especially for non-linear objective functions.
As discussed with @visr, users within the main network are allowed.
The text was updated successfully, but these errors were encountered: