AI-Based Predictive Control of Hybrid Offshore Floating Wind–Wave–Solar Systems for Resilient Microgrids

The increasing demand for sustainable and reliable offshore energy necessitates the integration of multiple renewable sources into resilient microgrids. This study addresses the limitations of standalone wind, wave, and floating solar systems, which exhibit high intermittency and volatility, compromising energy reliability and grid stability. To overcome these challenges, a hybrid offshore energy system is proposed, combining wind turbines, wave energy converters, and floating solar PV arrays. An AI-based predictive control framework, incorporating model predictive control and reinforcement learning, optimizes real-time energy dispatch, storage coordination, and demand tracking under variable environmental conditions. Simulation results demonstrate that standalone wind, wave, and solar generation fluctuated between 532.22–1332.80 kW, 210.83–701.00 kW, and 228.27–281.16 kW, respectively. In contrast, the integrated hybrid system achieved a smoothed total output ranging from 971.39 kW to 2314.95 kW, maintaining a mean operational capacity of 1567.33 kW, consistently exceeding peak load demands. The model predictive control minimized transient cost from 101.09 to 6.41, while reinforcement learning optimized cumulative policy rewards to 84.90. Economic evaluation revealed a competitive levelized cost of energy between 13.84–39.15 $/MWh, and environmental analysis confirmed greenhouse gas mitigation of 626.93 metric tons, surpassing single-source systems. The findings provide actionable insights for policymakers and stakeholders in designing resilient offshore hybrid microgrids that enhance energy security, reduce operational risks, and promote sustainable energy deployment.