Offshore wind energy has gained widespread attention and experienced a rapid development due to the significantly increasing demand on the renewable energy over the last few years. Currently, development of offshore floating wind turbine attracts lots of attentions to harvest more energy from sustained higher speed of offshore wind away from the coastline. Among many floating types, the spar type floating wind turbines, connected to the seabed through catenary mooring lines, are widely used in deep water. With stronger cyclic wind and wave loadings, the floating wind turbine could possibly experience severe fatigue damages at critical locations, which might lead to a catastrophic failure. Therefore, it is essential to evaluate the fatigue damage accumulation for the floating wind turbine during its entire lifetime (20-50 years). As demonstrated in the codes, specifications or design practices, the fatigue assessment requires massive computational costs and poses challenges of numerical simulations. Since structural dynamic responses are sensitive to complicated environmental conditions that are defined by many correlated or non-correlated random variables, defining an accurate statistical model for these environmental variables is an essential first step for the representative fatigue evaluation. Meanwhile, a thorough consideration of various load scenarios over the expected lifetime of the system is also necessary for evaluating fatigue damage accumulation over the entire life-cycle of the structure. In the present study, a copula-based multivariate probabilistic model is first built up to define the dependence among several wind and wave related environmental parameters. To achieve a faster convergence rate and reduce the number of required simulations, the Sobol sequence quasi-random sampling technique is implemented to select representative environmental conditions that could efficiently cover the complete design space. The selected environmental parameters are then utilized to characterize the wind and wave loads in the multi-physics code FAST to obtain the corresponding short-term structural dynamic response. The fatigue damage assessment is based on the rainflow counting method and the Miner's law with the Goodman correction. A surrogate model based on the Kriging framework is implemented to evaluate the short-term equivalent fatigue stress range under different environmental scenarios and at different locations of the floating wind turbine. In addition, the uncertainty effects originating from the complicated environmental conditions on the site will be incorporated into a probabilistic fatigue evaluation framework to assess the lifetime accumulated fatigue damage of a spar type floating wind turbine.
展开▼
