Effect of uncertainty sources on the reliability level of wind turbine rotor blades

Abstract

Wind turbine rotor blades are large composite structures performing most of their design life under random cycle loading patterns. Concurrently, material properties of the constituent plies exhibit inherent variability. In order to ensure a safe and costeffective design, uncertainty related to the basic variables (material properties, loads etc.) should be quantified and taken into account in the design calculations. Recently, the blade design standard DNVGL-ST-0376 was released, showing the trend of using dedicated probabilistic analysis for wind turbine blade design. A critical evaluation of the new edition of the standard is performed herein, particularly in terms of the ability of the suggested safety factors to satisfy the target failure probability level of 1E-04.To this end, probabilistic analysis methodology is employed, starting with the measurement uncertainty for the static and fatigue properties of the composite material and going all the way up to the blade failure probability on a layer by layer basis.The application is performed on the INNWIND.EU reference 10MW rotor blade of 90-m length. Furthermore, the current reliability level of the specific blade design, following the new standard DNVGL-ST-0376, is estimated considering the various failure modes, ie, fibre failure (short term and fatigue strength), buckling, and inter fibre failure of the composite laminates, while taking into account sources of variability that contribute to the physical, statistical, measurement, and model uncertainty. Results indicate that while for the static (extreme) analysis deterministic results are conservative, the opposite is observed for fatigue analysis. KEYWORDS composites, structural reliability, uncertainty analysis, wind turbine rotor blades 1 | INTRODUCTIONDesigning the 20 MW wind turbine of the future with approximately 250-m rotor diameter is a challenging task. The task becomes more complicated considering that wind turbine structures should continuously operate in a fully stochastic environment for their entire design life. To provide a safe and cost-effective design, uncertainty quantification and probabilistic analysis become mandatory.Reliability analysis of wind turbine rotor blades under extreme and variable amplitude loading has already been performed in the past. 1-5 The main effort, however, was put on the uncertainty quantification related to the environmental conditions, eg, wind speed and turbulence as well as the statistical treatment of the sectional stress resultants, with less emphasis placed on the uncertainty on the material properties. When mechanical properties were considered as random variables only part of the uncertainty was quantified; namely, the physical and statistical uncertainty related to the limited sample size of experimental tests. Measurement uncertainty was totally ignored. The available standards for composite materials do not provide any indication of the measurement uncertainty obtained through application of the experimental method, and therefore, this is difficult ...

Reliability model for wind turbine blade composites under alternate action of normal and extreme wind loads

“…Dimitrov et al 30 accounted for the loading direction and the randomness of material strength, and a spatial reliability analysis approach of wind turbine blade was presented. On the basis of probabilistic analysis methodology, Bacharoudis 31 investigated the effects of uncertainty sources on reliability of wind turbine blade. Hu et al 32 proposed a novel dynamic wind load model, and the probabilistic fatigue life of wind turbine blade was estimated.…”