Distributed dynamic feedback control for smart power networks with tree topology

In this paper, we design a distributed dynamic feedback controller that implements real-time economic optimization in power networks with tree topology. We consider the coupling between the dynamics of the power network and the appropriately formulated optimization problem in the real-time market, i.e., when the time-scale separation between markets and network dynamics fades. The design methodology is motivated by optimization decomposition methods in that the controller is derived from a primal-dual decomposition approach. We then prove the asymptotic stability of the overall (closed-loop) system in a scalable fashion. The performance of the controller is illustrated by numerical investigations.