Skip to Main Content
The future smart grid is expected to be an integration of intelligent microgrids featured by localized electricity generation, storage, and consumption. Wireless communication is a promising means to facilitate pervasive microgrid monitoring and control at a high flexibility and low deployment cost. In order to avoid a single point of failure, multiagent system (MAS) based decentralized microgrid control is widely considered. In this article, we present a consensus theory based multiagent coordination scheme for information discovery in microgrids via wireless networks. The information discovery process is fully distributed such that each agent only needs to communicate with its direct neighbors. The multiagent coordination is investigated in the presence of interference due to wireless transmissions. Specifically, the communication links among multiple agents may not be established simultaneously to avoid transmission collisions. In order to assure the accuracy of the information discovered by each agent, the convergence of the multiagent coordination is studied. We present the mechanism and principles for the multiagent coordination to achieve convergence in a microgrid via wireless network based on the characteristics of the information to be discovered. Three protocols with different complexities are proposed for normal microgrid operation, which can be readily implemented to off-the-shelf wireless communication devices. The protocols are further extended to achieve microgrid fault recovery by restricting the information exchange in a disconnected region to improve recovery speed. The performance of the proposed protocols is evaluated via a case study based on the network topology and configuration of a realistic microgrid test system. Open research issues and directions are outlined.