Distributed Operation Optimization of Active Distribution Network via a Sparse Communication Topology

In this paper, a distributed model predictive control framework is proposed to optimally operate active distribution networks under renewable energy and load uncertainty. It realizes coordinated adjustments of various devices in a distributed manner via a sparse communication topology. Prediction errors of photovoltaic output and load demand may lead to voltage violations and power flow variations. The active and reactive power of distributed resources are coordinated by their local agents to provide corrective control actions. In each control round, the agents are only responsible for collecting local data, controlling local devices, and iteratively exchanging disturbances and sensitivities with their neighboring agents, with no need of the knowledge of the whole network. Nevertheless, it has been verified that it can converge to the global optimal solution. The effectiveness of the proposed framework has been validated on two test systems.