A Scenario-Aware Supervisory Energy Management Framework for Grid-Interactive Photovoltaic-Battery Microgrids with Enhanced Operational Flexibility

Abstract

The decentralization of power networks and rising renewable penetration have intensified the need for microgrids that operate reliably in weak grids and under frequent disturbances. Conventional PV–BESS (photovoltaic-battery energy storage system) control relies on a fixed grid-forming (GFM) or grid-following (GFL) mode, and is therefore prone to instability during transitions, inefficient energy utilization, and poor resilience to faults and rapid load or generation variations. This work proposes a dual-mode coordinated control architecture for a grid-interactive PV–BESS microgrid that switches autonomously between grid-forming and grid-following operation according to real-time system conditions. An adaptive energy management strategy monitors five operational scenarios: irradiance fluctuations, sudden load changes, black-start operation, islanding transitions, and fault ride-through events. Droop-based GFM control, PQ-regulated GFL support, and a scenario-aware supervisory layer together deliver smooth mode transitions, improved voltage and frequency regulation, and efficient PV–BESS coordination without external intervention. The contribution lies not in the individual control primitives, which are established, but in three couplings between them: a continuous blending law that makes the GFM–GFL transition a designed trajectory rather than a discrete switch; inertia and damping adaptation closed on an online grid-strength estimate, with a Lyapunov stability envelope; and a single arbitration ordering that resolves conflicting objectives when disturbances overlap. Electromagnetic-transient simulations built on switching-function converter models and a second-order electrochemical battery model demonstrate stable behavior under weak-grid conditions at a short-circuit ratio (SCR) of 0.44, with post-disturbance frequency restored inside the supervisory band, transient excursions contained within the grid-code ride-through envelope, and substantially faster grid resynchronization than fixed-mode approaches. PV utilization remains high with reduced curtailment, and unnecessary BESS cycling is avoided. Validation is entirely simulation-based, with staged hardware-in-the-loop testing defined as the next step, and the framework offers a scalable pathway toward resilient, adaptive renewable microgrids.