On the development and application of short design events for the prediction of extreme responses of floating offshore wind turbines

• A method for producing response-conditioned wave profiles is applied to wind for the first time. • A method for short design events using response-conditioned wave and wind profiles is developed, with an example. • OpenFAST models of eight floating wind concepts covering the four main platform types are tested against design standards. • Single design events estimate to within − 19 % to +20 % of the traditional 1-h simulation approach. • For responses with limited higher-order effects, single design events estimate within − 6 % to +20 % of the traditional approach. • Where 2nd order effects are important, constrained events perform better, within − 5 % to +20 % of the traditional approach. • Short design events have potential in early design stages to explore a wide range of design conditions. For the prediction of design loads of floating wind turbines in extreme sea states, design standards recommend simulating multiple irregular wave and turbulent wind time series of one to three hour duration. For fixed offshore wind turbines, a constrained wave approach with 10-min simulations is permitted, significantly reducing simulation times. This method is not recommended for floating turbines as the largest wave does not always cause the largest response of interest. This work seeks to develop a response-conditioned, short design event approach to create an equivalent constrained wave method for floating turbines. The responses of spar, semi-submersible, tension leg platform, and barge-type floating wind devices under extreme conditions are modelled numerically using OpenFAST. A method for constraining short response-conditioned wind and wave time series is developed. A comparison of extreme pitch, nacelle accelerations, mooring loads, and tower base bending moments using irregular waves and constrained and unconstrained response-conditioned events is given. The constrained short time series efficiently produced extreme responses within − 5 % to + 20 % of those from extended irregular wave and wind series. Care is needed when using unconstrained time series in situations where higher-order effects dominate, such as surge or tower natural periods excited by second-order wave loading.