Meta-models for fatigue damage estimation of offshore wind turbines jacket substructures

Abstract

One of the main structural components of offshore wind turbines is the substructure which bridges the gap between seabed and tower foot. One possible concept employed in intermediate water depths for turbines with high-rated power is the jacket. This structure is excited by several environmental impacts like wind and wave loads or centrifugal loads from the rotor motion. In order to reach competitive costs of energy, it is crucial to minimize the lifetime capital expenses by means of robust and reliability-based design. However, a simulation-based optimization approach on the full scale model requires high numerical capacity. In this work, the problem of numerically expensive fatigue life evaluation is addressed by the utilization of a meta-model approach. The performance of two meta-models solutions, namely Kriging and Interval Predictor Model, is compared. In particular, the different behavior of the probabilistic confidence intervals of the Kriging regression and the interval bounds of the IPM is discussed.