Marine-adapted hybrid algorithm-driven multi-objective optimization for offshore island integrated energy systems with underwater compressed air energy storage

Offshore island integrated energy systems (OIIES) are critical for harnessing offshore renewable energy. Leveraging deep-sea hydrostatic pressure gradients, underwater compressed air energy storage (UWCAES) presents an optimal energy solution for marine islands. However, research on UWCAES integration within OIIES remains limited, hindering system deployment and marine energy utilization. Herein, a UWCAES-based OIIES framework and its mathematical model are proposed, and a special operational optimization strategy is established. Moreover, a novel hybrid probability density function is also proposed, characterizing wave energy generation uncertainty, and is validated with historical power data. To address the tendency of the non-dominated sorting genetic algorithm II (NSGA-II) toward local optima, the marine predator algorithm (MPA) is integrated, resulting in the development of MPA-NSGA-II to enhance exploration capability. Algorithm validation employs three benchmark scenarios; comparative analysis demonstrates a 16.99% reduction in economic cost and a 7.36% increase in energy utilization compared to hydrogen storage systems. Relative to regular battery-only systems, the proposed UWCAES-based OIIES achieves cost savings of 30.49% and a utilization increase in 2.01%, confirming both algorithmic efficacy and system superiority. This study provides a theoretical foundation for optimizing UWCAES-based OIIES, advancing marine resource development and exploration.