Development Of an AI-Driven Supercapacitor-Integrated Control System for Real-Time Mitigation of Power Fluctuations in Hybrid Solar-Wind Energy Networks

The increasing penetration of hybrid solar-wind energy systems into modern electric power networks is fundamentally constrained by rapid and stochastic power fluctuations that degrade voltage stability, power quality, and system reliability. This study focused on an AI-driven supercapacitor-integrated control system for real-time mitigation of power fluctuations in hybrid solar-wind energy systems. A comprehensive modeling architecture was developed, incorporating photovoltaic and wind power generation, DC-link dynamics and a supercapacitor-based hybrid energy storage system. An artificial intelligence control strategy, formulated as a Markov Decision Process, was deployed to dynamically regulate supercapacitor charge-discharge activities under varying environmental and load conditions. The developed system was validated using high-fidelity time-domain simulations and hardware-in-the-loop (HIL) testing to assess real-time feasibility. Results demonstrated effective stabilization of the DC-link voltage within permissible limits, millisecond-scale control latency and well-regulated supercapacitor voltage and state-of-charge dynamics under severe power fluctuations. The system achieved a high overall efficiency (92.5%), low voltage harmonic distortion (3.1%), competitive energy cost (0.11 USD/kWh), and substantial annual CO₂ emission reduction (1450 kg). These findings confirm that integrating fast-acting supercapacitor storage with AI-based control significantly enhances power quality, stability, and sustainability in hybrid renewable systems. It is recommended that future work focus on large-scale field deployment and extension to multi-layer energy storage architectures to further improve grid resilience and renewable hosting capacity.