AI-Based Optimization and Hydrodynamic Performance of Floating Pontoon Breakwaters through Experimental and Numerical Modeling

Ensuring effective coastal protection requires not only attenuating incoming wave energy but also preserving the operational reliability of marine infrastructure. This study investigates the hydrodynamic behavior of two pontoon-type floating breakwaters—simple and stepped—through a dual approach combining controlled wave flume experiments with advanced numerical modeling in ANSYS AQWA. A scaled physical model (1:15) was subjected to regular wave conditions, with incident wave heights ranging from 2.2 cm to 3.3 cm and wave periods between 0.44 s and 0.84 s. The experimental results revealed that the simple pontoon structure achieved optimal energy dissipation at shorter wave periods (0.44 s), whereas the stepped configuration performed more efficiently under longer wave conditions (0.84 s). The transmission coefficient (Kₜ) ranged from 0.2 to 1.5, with the stepped design consistently demonstrating enhanced wave energy reduction. Numerical simulations closely mirrored the experimental outcomes, exhibiting a maximum deviation of only 8%. To improve breakwater performance, we deployed a suite of artificial intelligence (AI) methods: artificial neural networks (ANNs) to forecast wave-induced loads; genetic algorithms (GAs) to explore the design space and tune structural parameters; and reinforcement learning (RL) to enable adaptive, real-time control. In combination, these techniques reduced the transmission coefficient from 0.52 to 0.40 and, notably, increased computational efficiency by 88%. A digital-twin environment, instrumented with an IoT sensor network, enabled continuous, real-time appraisal of structural condition. The results underscore the primacy of breakwater geometry and draft configuration and show that AI-guided optimization materially strengthens wave-attenuation capacity—yielding a scalable, cost-efficient pathway for resilient coastal infrastructure.