Dear Prof. Zimmerman, I'm currently using running DC power flows in Matpower. Alternatively, I compute the shift factors by myself. As far as I'm concerned, for the same set of power injections and withdrawals, both techniques should provide me identical flows over my network, but surprisingly this is not what is happening. Even though I get different power flows, I've observed that the net export (flows from all lines connected to a bus) for all nodes are the same in both mechanisms.
The shift factors are apparently well computed. Does this make any sense? I would strongly appreciate if you could provide some feedback on this. Kindest regards, Jose
