Coordinated Distributed Predictive Control for Voltage Regulation of DC Microgrids with Communication Delays and Data Loss

This paper is concerned with the voltage tracking problem of DC microgrids subject to communication delays and packet losses, for which existing work commonly adopts passive fault-tolerant approaches. To accurately compensate for the above communication constraints experienced by DC microgrids, a coupled discrete microgrid model is developed on the basis of the physical laws of actual DC microgrids. Unlike the practice of passively tolerating communication delays, through the physical model established, this paper suggests a consensus-based proportional-integral predictive control strategy, which can actively compensate for communication delays and consecutive packet dropouts encountered by DC microgrids. Under the observer-based distributed networked predictive control framework, each distributed generation subsystem in the microgrid exchanges its own measurements over the network and integrates information from controllers of other units to achieve output voltage consensus in the presence of time delays and packet losses. Furthermore, to demonstrate the generality of the proposed method, the sufficient and necessary conditions for the DC microgrid system to accomplish voltage tracking are given. These conditions are rendered in the form of matrix eigenvalues associated with the physical connections and communication couplings between distributed generation units. Besides, the stability and convergence of the closed-loop microgrid system are given in the form of linear matrix inequality based on the Lyapunov function. Finally, the performance of the proposed control scheme is evaluated in terms of its convergence, robustness to load variations, and plug-and-play functionality through the built photovoltaic cell-based (with battery banks) DC microgrid hardware system.