Posters

Abstract

Mooring line integrity is a critical concern for floating offshore wind platforms, where the failure of a single line can compromise station-keeping, increase fatigue on remaining moorings, interrupt electricity production, and in extreme cases lead to catastrophic loss of the structure. Traditional monitoring approaches rely on dedicated line sensors such as load cells and strain gauges, which add cost, complexity, and maintenance requirements in harsh offshore conditions. For large floating wind farms, with dozens of mooring lines to be monitored continuously, a more scalable solution is required. This study presents a machine learning–based methodology for detecting mooring line failures using only standard turbine measurements: position data and nacelle wind speed and direction. The method builds on prior work with floating production, storage, and offloading (FPSO) systems, but is here reframed for floating wind applications. The underlying principle is that a failed line produces an unbalanced restoring force, shifting the floater’s equilibrium position relative to intact conditions. This effect is more pronounced in floating wind turbines with non-redundant mooring systems, where failure is expected to cause a large excursion. Although platform dynamics and wind controller effects introduce additional uncertainty, the large excursions under failure dominate, ensuring that detection remains feasible and reliable.