Reliability assessment of a 15 MW floating offshore wind turbine considering climate change

This study evaluates the reliability of a 15 MW floating offshore wind turbine (FOWT) under future climate conditions, considering the correlation of failure modes. An error-controlled multi-input-multi-output surrogate model based on support vector regression is constructed for the FOWT, propagating uncertainties across multiple structural parameters while enhancing computational efficiency. Based on the trained surrogate model, Monte Carlo simulations are performed to obtain the marginal and joint probability distributions of key responses, thereby capturing the correlations among failure modes. Additionally, future wind data for three scenarios (SSP126, SSP245, and SSP585) are collected and calibrated point by point. By integrating historical wind-wave joint probability distributions and projected wind data for future climates, the wind-wave joint probability distributions under future environmental conditions are derived to calculate the failure probabilities of FOWTs located along the Southeast coast of China. Results demonstrate that different emission scenarios apparently influence the failure probability, and there exists a significant discrepancy in the spatial distribution of the FOWT’s failure probability.