A Data-Driven Centralized Secondary Control Design Methodology for Microgrids

Microgrids play an important role in modern power systems, integrating distributed energy resources (DER) that enhance resilience through advanced control capabilities. Managing frequency, voltage, and power dynamics in microgrids under varying conditions, however, poses significant challenges. This paper proposes an adaptive, data-driven secondary control framework for real-time tuning of hierarchical controllers in both islanded and grid-connected modes. The approach employs a multiple-input multiple-output (MIMO) Virtual Reference Feedback Tuning (VRFT) method with a novel low-memory formulation optimized for digital signal processor (DSP) implementation. Integrated with a performance detector, it enables automated controller retuning during open- and closed-loop operations, ensuring robust reference tracking and dynamic performance under disturbances. Hardware-in-the-loop (HIL) simulations, incorporating significant communication delays, validate the framework’s effectiveness. The results highlight the proposed method’s practical applicability, offering a scalable solution for microgrid management in real-world conditions.