Presentations

Abstract

Floating wind structures are exposed to highly dynamic metocean conditions that accelerate fatigue, raise failure risks, and increase operational costs. Current predictive maintenance strategies rely heavily on computationally expensive simulations, which are unsuitable for real-time operation at farm scale. The ePROA project addresses this challenge through the development of a digital twin powered by surrogate models, enabling accurate and fast prediction of structural responses. Surrogate models, including Fourier Neural Operators (FNOs) and physics-informed neural networks, were trained on extensive datasets generated with OrcaFlex and AQWA under diverse sea states. These models replace time-intensive simulations with efficient approximations, allowing prediction of mooring line tensions, tower bending moments, and fatigue indicators in near real-time. The system incorporates anomaly detection, remaining useful life (RUL) estimation, and decision-support dashboards via Grafana, offering operators clear, interpretable insights. By enabling proactive interventions, this approach reduces unplanned maintenance, lowers costs, and increases turbine availability. Beyond cost savings, predictive maintenance driven by AI-enhanced digital twins improves safety, asset lifespan, and resilience of offshore wind farms, making floating wind technology more competitive. This research demonstrates the potential of surrogate models to transform monitoring and maintenance, establishing a new paradigm for reliable, sustainable offshore operations.