Secondary control for DC microgrids with optimal sparse feedback

This paper designs a sparse controller to conduct DC microgrid secondary control. There is a considerable amount of literature on centralized, distributed, and decentralized secondary controllers for DC microgrids. A decentralized controller only utilizes local information to make decisions, hence often cannot accomplish system-wise regulation goals; on the other hand, existing centralized and distributed controllers are known to have strong dependence on active information exchanges hinging on dense communication networks. This dependence can induce heavy computational burden as well as vulnerabilities to cyber incidents. To reduce this dependence while achieving system-wise control objectives, a sparse secondary control scheme is designed by optimally reducing the communication complexity. We propose a novel algorithm to identify redundant state feedbacks whose impact on the control performance is marginal. Removing redundant feedbacks effectively assists us to design an optimal control structure, with communication sparsity and control performance guaranteed. The effectiveness of the proposed work is demonstrated using case studies.