Time-domain uncoupled analyses for seismic assessment of land-based wind turbines
Time-domain uncoupled analyses for seismic assessment of land-based wind turbines
复制标题
DOI：
10.1016/j.engstruct.2016.05.043
复制
发表时间：
2016-09
影响因子：
5.5
通讯作者：
F. Santangelo;G. Failla;A. Santini;F. Arena
中科院分区：
文献类型：
--
作者：
F. Santangelo;G. Failla;A. Santini;F. Arena
地震分析在地震危险区域的陆基和海上风力涡轮机设计中起着重要作用。对于地震评估，国际标准和指南允许结合两个单独的分析，一个在风下，另一个在地震下，作为计算昂贵的，完全耦合的时域模拟的替代方案。在这些非耦合分析中，通常使用标准加速度反应谱计算单独的地震反应，包括称为空气动力阻尼的附加阻尼。然而，通过响应谱方法，重要的非线性源，如与基础的灵活性，不能正确占。专注于陆基风力涡轮机，本文探讨了一个时域实现解耦分析，这可能涉及到一个非线性基础模型。该案例研究是一台5 MW基线风力涡轮机，位于由非线性弹簧建模的桩基础上。对于不同的地震记录和风速，与完全耦合模拟的比较表明，在时域中实现的解耦分析的组合产生准确的结果，提供了一个适当的水平的气动阻尼包括在模型中。值得注意的是，可以看出，这样的气动阻尼水平与一般建议的响应谱为基础的解耦分析。
Seismic analysis plays an important role in the design of land-based and offshore wind turbines in areas at seismic hazard. For seismic assessment, International Standards and Guidelines allow combining two separate analyses, one under wind and another under earthquake only, as alternative to computationally expensive, fully-coupled time-domain simulations. In these uncoupled analyses, the separate earthquake response is generally computed using the standard acceleration response spectrum, upon including an additional damping referred to asaerodynamicdamping. By a response-spectrum approach, however, important sources of nonlinearity, such as those related to foundation flexibility, cannot be properly accounted for.Focusing on land-based wind turbines, this paper investigates a time-domain implementation of uncoupled analyses, which may involve a nonlinear foundation model. The case study is a 5 MW baseline wind turbine, resting on a pile foundation modeled by nonlinear springs. For different earthquake records and wind velocities, comparisons with fully-coupled simulations show that the combination of uncoupled analyses implemented in the time domain yields accurate results, provided that an appropriate level of aerodynamic damping is included in the model. Notably, it is seen that such aerodynamic damping level agrees with the one generally recommended for response-spectrum based uncoupled analyses.