Bivariate reliability analysis for floating wind turbines
Oleg Gaidai,Vladimir Yakimov,Fang Wang,Jiayao Sun,Kelin Wang
DOI: https://doi.org/10.1093/ijlct/ctad108
IF: 2.3
2024-01-01
International Journal of Low-Carbon Technologies
Abstract:Abstract Wind turbines are designed to withstand extreme wind- and wave-induced loads, hence a reliability study is vital. This study presents a bivariate reliability approach, suitable for accurate assessment of critical forces and moments, occurring within the wind turbine’s critical mechanical parts, such as the drivetrain. A ecently developed bivariate modified Weibull method has been utilized in this study. Multivariate statistical analysis is more appropriate than a univariate one, as it accounts for cross-correlations between different system components. This study employed a bivariate modified Weibull method to estimate extreme operational loads acting on a 10-mega watt (MW) semi-submersible type floating wind turbine (FWT). Longitudinal, bending, twisting, and cyclic loads being among typical load types that FWTs and associated parts are susceptible to. Furthermore, environmental loads acting on an operating FWT being impacted by incoming wind’s stochastic behavior in terms of wind speed, direction, shear, vorticity, necessitates accurate nonlinear extreme load analysis for FWT critical parts such as the drivetrain. Appropriate numerical methods were used in this study to model dynamic, structural, aerodynamic, and control aspects of the FWT system. Bending moments acting on the FWT drivetrain have been obtained from SIMPACK (Multibody Simulation Method), given realistic in-situ environmental conditions. For a 5-year return period of interest, a bivariate modified Weibull method offered robust assessment of FWT’s coupled drivetrain’s bending moments.
energy & fuels,thermodynamics