Lifetime load sensitivity analysis of offshore wind turbine

In the design of offshore wind turbines (OWT), industrial practice simplifies the environmental design conditions in accordance with the standard guidelines, which are based on a limited set of Design Load Case's (DLC). This deterministic design strategy requires the use of partial safety factors (PSF) to account for uncertainty in the design process - since full probabilistic analysis is computationally expensive - which builds in conservatism. During operation, lighter lifetime loads due to uncertainties such as those in the information on the site-specific conditions and milder utilization pre-construction design stage, highlight an opportunity to evaluate and quantify the reserve life introduced in the conservative design practices. This study proposed a computational strategy to quantify the effect of environmental parameters in the design loads of OWT in North-Sea environmental conditions. Among these parameters uncertainty associated with the turbulence intensity, wave peak period have been propagated in the support's Damage Equivalent Loads (DEL), whereas mean wind speed, wave height and wind-wave misalignment considered as global parameters. Three sensitivity analysis techniques such as PAWN, Delta, and standardized rank regression coefficients (SRRC) are used to evaluate the total effect of the five key parameters on the DELs of IEA's 15 MW reference OWT. This analysis is conducted based on a proposed improved semi-probabilistic design model technique, which uses selective sampling to reduce the computational restraints of the OWTs environmental design space. The models resulting analysis shows a high impact of mean wind speed and wave height, however, on a local sub-categorized case, in different operational wind speeds, the turbulence intensity and wave peak period cause high variability in results, which is largely unrepresented in the current industry design models.

Keywords: Offshore wind turbine, Uncertainty Quantification, Sensitivity, Fatigue life, Structural Reliability, Remaining Useful Life.