Experimental study of distributed control for renewable energy/energy storage/hydrogen system-based microgrid clusters using Raspberry Pi agents

Abstract

Microgrid clusters (MGCs) are emerging as a promising solution for integrating diverse DC and AC technologies, thereby enhancing flexibility and resilience. This research presents an experimental study of a control strategy for a MGC comprising two microgrids (MGs), interconnected with each other: a DC MG with a wind generator, an ultracapacitor, DC local loads, and a hydrogen system; and an AC MG that includes an electric battery, AC local loads, and a photovoltaic power plant. Through experimental validation, this research produces valuable findings into the design and implementation of advanced MGC technologies. The proposed control strategy leverages local controllers and a distributed control architecture that includes two control agents, which coordinates the power management between the technologies of the MGC. The experimental setup, consisting of an OPAL-RT unit and two Raspberry Pi boards, is implemented for the real-time testing and verification of the control methodology across a range of operational scenarios. The real-time experimental results validate the effectiveness of the proposed control methodology. Additionally, the control strategy exhibits superior integral time absolute error performance compared to a centralized control architecture. Finally, a study on communication delays across all control agents reveals that while performance degrades, system stability is maintained.