This paper presents a methodology to develop machine learning models designed to predict, in real-time, hourly fatigue damage accumulation in the mooring lines of floating wind turbines, from measurements of five environmental variables: wind speed, wind direction, wave height, wave period, and wind-wave misalignment. The proposed tool is intended for predictive maintenance applications, which has been identified as a key area for cost reduction in floating wind.

Floating offshore wind technology has high potential but still faces challenges for gaining economic competitiveness to allow commercial market uptake. Hence, design optimization plays a key role; however, the final optimum floater obtained highly depends on the specified optimization problem. Thus, by considering alternative structural realization approaches, not very stringent limitations on the structure and dimensions are required. This way, more innovative floater designs can be captured.

One particular problem with structures operating in seas is the so-called fatigue phenomenon. Cyclic loads imposed by waves and winds can cause structural failure after a number of cycles. Traditional methods have some limitations. This paper presents a developed design framework based on fracture mechanics for offshore wind turbine support structures which enables design engineers to maximise the use of available inspection capabilities and optimise the design and inspection, simultaneously.