Load condition determination for efficient fatigue analysis of floating offshore wind turbines using a GF-discrepancy-based point selection method

Fatigue failure is one of the decisive factors to be considered in the structural safety design of floating offshore wind turbines (FOWTs). In the fatigue assessment of FOWTs, adequate simulations are required to account for the variability of long-term environmental load conditions. Although the number of potential environmental conditions can be infinite, only a limited number of representative load conditions are used in practice. This inevitably introduces uncertainty into the fatigue estimation, which is usually referred to as the finite sampling -induced uncertainty. To reduce this uncertainty, numerous simulations are required at the expense of unbearable computation burdens. The present study proposes a method to determine the representative load conditions for efficient fatigue evaluation of FOWTs based on the GF-discrepancy of point set. The copula method is adopted to establish the joint probabilistic model of long-term wind and wave conditions. The in-house developed Sto-DRAFOWT model and the SN-curve approach are employed for the fatigue evaluation. The fatigue uncertainty due to finite sampling under different load condition selection methods is then assessed. The comparison results indicate that the proposed method can significantly reduce the uncertainty in the fatigue estimation of FOWTs against conventional methods, thus improving the fatigue analysis efficiency of FOWTs.