摘要

This paper proposes a decentralized approach for frequency control and economic dispatch in smart grids with distributed energy resources and responsive demands. The approach is implemented based on a distributed multi-agent system, in which each agent works in a collaborative manner via local computation and peer to peer communication. The approach is comprised of two strategies. A consensus-based strategy is implemented at the supply side to regulate system frequency economically. Self-adaptive updating rules are proposed for the key parameters to improve the accuracy and robustness of the consensus-based strategy. A threshold-based strategy is implemented at the demand side to provide ancillary services for frequency regulation. Simulation results are presented to demonstrate the effectiveness and robustness of the approach.