Distributed Secondary Control of DC Microgrids Under Unreliable Communication Networks

This paper investigates the voltage restoration and optimal load sharing problem of direct-current (DC) microgrids with considering random switching topologies and communication delays. The topological random switch is driven by a continuous-time Markov process with its state corresponding to all communication topologies among distributed energy resources (DERs). A distributed communication-based secondary control protocol is proposed to achieve voltage restoration across multiple DERs while optimizing load distribution for cost-effective operation in the mean square sense. Such control protocol is implemented in a fully distributed manner, with each controller only using the local information and the information from its neighbors, thereby enhancing its robustness, flexibility, and scalability. A sufficient condition is given to ensure the mean square exponential stability of the stochastic time-varying delayed systems by constructing a novel topology-dependent degenerate Lyapunov functional. Finally, the effectiveness and robustness of the proposed control protocol are verified by comparing with the random switching topologies and communication delays based on a set of simulation examples.