Chapter37 Including SN-Curve Uncertainty in Fatigue Reliability Analyses of Wind Turbines Jennifer M. Rinker and Henri P. Gavin Abstract The reliability of wind turbines with respect to fatigue loads is affected by uncertainties in dynamic load effects and the material’s resistance to the fatigue loads. The design standard published by the International Electrotechnical Commission (IEC) specifies partial safety factors to account for these uncertainties and is intended to produce systems that are reliable with respect to fatigue loading. This paper presents a method to directly propagate the uncertainty in both the dynamic response and the fatigue resistance of a component through to the subsequent accumulated damage. Damage indices are calculated using both the proposed method and the IEC safety factors, providing a comparison of the two uncertainty quantification methods. Keywords Wind turbine • Fatigue • Reliability • SN curve • Uncertainty 37.1 Introduction To be a factor in the energy market, the cost of energy (COE) of a particular energy source must be low enough to be competitive with the other sources of energy that are already available. The U.S. government has set a goal to have 20 % of the U.S. energy portfolio coming from wind in 2030 [1], but one of the current limitations of wind energy is that the level of uncertainty in profits is high, which raises the COE and makes potential investors wary of developing the market [2]. In places with little unpopulated land such as Europe, wind energy is now transitioning from onshore to offshore turbines, and the high costs of maintenance and repair for offshore sites will only increase the importance of wind turbine reliability in the future. Currently, the main document that governs wind turbine design world-wide is the design standard published by the International Electrotechnical Commission (IEC) [3]. This document contains both design guidelines and recommended protocols for many different aspects of wind turbine design, construction, and erection. The structural design of wind turbines is separated into load cases, each corresponding to a different operational situation or type of loading, and a design should satisfy all of the load cases that are deemed relevant to the design. The design for fatigue involves large uncertainties in both fatigue-inducing dynamic responses and in material fatigue resistance. Fatigue failures are especially difficult to predict because fatigue accumulates over periods of time that are longer (in terms of stress cycles) than can be physically tested. Failure of parts due to fatigue can cause a significant increase in the COE due to downtime and maintenance costs, so it is important to design a turbine that can withstand environmental loadings; however, over-designing a piece will unnecessarily increase its cost. Thus it is important to design a piece such that it is adequately strong, but no stronger. Uncertainties in the expected maintenance and downtime costs of large wind farms (and the associated COE) depend significantly on the likelihood of fatigue failures. The IEC method recommends the use of partial safety factors (PSFs) for a margin of safety in design, and most literature that has been published in the last decade follows this method of design. Veldkamp summarized the PSF design methodology [4] and also noted a way to calibrate the PSFs to reduce the level of over-design [5]. Sörenson et al. used the PSFs in their analysis of the IEC’s effective turbulence model [6]. Some previous studies, however, did not use the PSF methodology, J.M. Rinker ( ) • H.P. Gavin Civil & Environmental Engineering, Duke University, Hudson Hall, Box 90287, Durham, NC 27708, USA e-mail: jennifer.rinker@duke.edu; hpgavin@duke.edu H.S. Atamturktur et al. (eds.), Model Validation and Uncertainty Quantification, Volume 3: Proceedings of the 32nd IMAC, A Conference and Exposition on Structural Dynamics, 2014, Conference Proceedings of the Society for Experimental Mechanics Series, DOI 10.1007/978-3-319-04552-8__37, © The Society for Experimental Mechanics, Inc. 2014 375
RkJQdWJsaXNoZXIy MTMzNzEzMQ==