Distributed noise-resilient economic dispatch strategy for islanded microgrids

Abstract

Economic dispatch (ED) plays an important role in economic operation of the microgrid (MG). However, the communication links among distributed generators (DGs) may practically be corrupted by additive noise, resulting in erroneous deviations from ED commands. To deal with such issues, this study proposes a distributed noise-resilient ED strategy for islanded MG. Owing to consensus-based implementation, the proposed ED strategy is implemented in a fully distributed manner, which enables peer-to-peer communication among DGs without the necessity of a central controller. In addition, the proposed ED strategy involves post-iterate averaging technique to further enhance its convergence under additive communication noise. In this way, different from the existing ED strategies, the proposed ED strategy is fully distributed and resilient to the communication noise. Furthermore, the effectiveness of the proposed ED strategy is evaluated on an islanded MG, where the communication links are corrupted by different levels of additive noise. Finally, in order to fully take account of the uncertainty and stochastic nature of different operation scenarios, the Monte-Carlo simulations are carried out, of which the simulation results demonstrate that the proposed ED strategy is superior to the existing ED strategies, in terms of the convergence property under additive communication noise.