Sensitivity analysis of wind turbine fatigue reliability: effects of design turbulence and the Wöhler exponent

Abstract. Fatigue reliability assessment of wind turbine components involves three major sources of uncertainty: material resistance, loads, and the mathematical models that connect the other two. Many studies focus on decreasing the uncertainty in fatigue load assessments via different approaches including more detailed characterization of the turbulence standard deviation (turbulence) in the design phase. The IEC standard suggests two different distributions in Edition 3 and Edition 4 as alternatives for the representative turbulence in its Normal Turbulence Model (NTM): lognormal and Weibull. There are debates on whether the two suggested distributions provide adequately safe reliability in relevant load channels since the established design safety factors are calibrated based on the representative turbulence approach. The current study addresses the debate by comparing annual reliability based on different scenarios of NTM using a probabilistic approach. More importantly, it elaborates on the level of importance of this matter or most of the other concerns about the load’s uncertainty in the context of reliability considering all uncertainty in the material properties and the linear damage accumulation rule.

Using aeroelastic simulations of the DTU 10MW reference wind turbine and a simple model for assessing fatigue reliability based on the distribution of damage equivalent load (DEL), we study the changes and the trends in the annual reliability as well as the relative importance of each of the uncertainty groups. In addition, we investigate the effects of different fatigue exponents on the overall results in the blade root flapwise and the tower base fore-aft load channels.

The results show that generally using the distribution of turbulence in each mean wind speed results in less conservative annual reliability levels compared to representative turbulence. The difference in the reliability levels varies according to turbulence distribution and the fatigue exponent. In the case of the tower base, the difference in reliability after 20 years can be up to 50 %. However, the model and material uncertainty have much higher effects on the reliability levels compared to DEL uncertainty. Knowledge about such differences in the reliability levels due to the choice of turbulence characterization method is especially important as it impacts the extent of lifetime extension through reliability re-assessments.