<![CDATA[This study examines collective frequency desynchronization in artificial intelligence-stabilized renewable energy networks and its effect on autonomous grid recovery capacity. A calibrated digital-twin model was developed for a representative islanded 20-kV hybrid renewable distribution network in East Nusa Tenggara, Indonesia, consisting of 8.5 MWp solar photovoltaic generation, 2.4 MW wind generation, a 4 MW/16 MWh battery energy storage system, 18 distribution buses, and a peak load of 10.2 MW. The model was calibrated using 30 days of operational profiles, including renewable power output, feeder load, battery state of charge, inverter response, and local frequency measurements. Dynamic time-domain simulations were conducted under 1,500 disturbance scenarios involving renewable intermittency, sudden load variation, battery power limitation, communication delay, and inverter control mismatch. The results show that AI-based frequency stabilization improved recovery performance under normal communication conditions, reducing average frequency recovery time from 11.2 s to 8.7 s. However, when renewable penetration exceeded 70% and communication latency increased above 200 ms, coordinated AI control actions produced phase-shifted responses among inverter clusters, causing partial frequency incoherence. Under this condition, the autonomous recovery capacity index declined from 0.86 to 0.64, while average recovery time increased to 13.9 s. These findings indicate that AI-based stabilization can strengthen renewable grid resilience, but only when latency-aware coordination and frequency-coherence constraints are integrated into the control architecture. The study provides a practical simulation framework for assessing autonomous recovery capability in low-inertia, inverter-dominated renewable energy networks]]>
Copyrights © 2026
