Description

Due to the continuous search for cost reductions in wind energy, probabilistic design has become a hot topic. Data from tests, measurement campaigns, and operation can be utilized to reduce uncertainties, and with probabilistic design methods, this can be translated into a less conservative design. This is more optimal both from an economic and a sustainability perspective. When ed. 4 of the main design standard for wind turbines IEC 61400-1 was published in 2019, a reliability target was specified, which opened for the possibility of probabilistic design [1]. To ensure consistency and acceptance across the industry, the development of a technical specification on probabilistic design (IEC TS 61400-9 [2]) was initiated in 2021. In IEC 61400-1, the annual target for the reliability index was specified as β=3.3 for component class 2 (annual failure probability 5⋅10^(-4)) based on ISO2394 [3] and the existing safety factors. However, the same safety factors are used for a wider range of load cases, components, and failure modes, and some variation of the obtained reliability level is generally expected. This work deals with the issue of how to formulate more elaborated reliability requirements for probabilistic design. It should be in consistence with IEC 61400-1, but discrepancies are found between the stated reliability target and the reliability level implicitly given through the safety factors. If the implicit reliability level for some cases is less than the stated target, probabilistic design will often lead to a more expensive design. However, in some cases a reduced reliability level can be motivated by a risk-based assessment [4], which is also the background for the targets in ISO2394. In this work, the reliability level is estimated for various load cases in IEC 61400-1, and a risk-based assessment of the reliability is performed. Typically, the target reliability is interpreted as the minimum reliability in the worst year. It is found that a risk-based assessment can motive differentiation of the reliability requirements between load cases based on the development of the reliability over time. This is in line with the opinion of the original author of the reliability indexes in ISO2394, who argued that the target should be applied to the asymptotic value of the renewal process [5], which approximately corresponds to the reliability corresponding to the average annual failure probability over a fixed lifetime. But this raises the question of how to formulate this to ensure rational through life integrity management.

References:
1. IEC 61400-1 ed. 4 Wind energy generation systems - Part 1: Design Requirements 2019.
2. IEC CD TS 61400-9 Wind energy generation systems - Part 9: Probabilistic design measures for wind turbines 2022.
3. ISO2394 General principles on reliability for structures; International Organization for Standardization, 2015;
4. Nielsen, J.S.; Sørensen, J.D. Risk-based derivation of target reliability levels for life extension of wind turbine structural components. Wind Energy 2021, 1–18, doi:10.1002/we.2610.
5. Rackwitz, R. Optimization — the basis of code-making and reliability verification. Struct. Saf. 2000, 22, 27–60, doi:10.1016/S0167-4730(99)00037-5.

Period25 May 2023
Event titleWind Energy Science Conference 2023
Event typeConference
LocationGlasgow, United KingdomShow on map
Degree of RecognitionInternational