基于桩土耦合作用的单桩式风机承载力可靠性研究

Reliability study on bearing capacity of monopile wind turbines based on structure-soil coupling

大直径单桩是海上风机重要的基础型式,其承载力可靠性是桩基设计的关键问题。本文考虑实际多土层的地质条件以及风速、波高、周期等环境参数的相关性,改进并采用BP神经网络与蒙特卡罗模拟相结合的方法,对大直径单桩风机在正常使用极限状态(SLS)下的承载力可靠性开展研究;基于桩-土接触面模型进行确定性的承载力分析,分别对砂土、粘土海床中的准确性进行验证,并采用该模型确定神经网络训练点的准确解;最后,以LW 8MW风机为例,进行SLS下单桩风机承载力可靠性分析。该方法可以为国内后续海上风电场的设计建设提供参考。

Large-diameter monopiles are an important type of offshore wind turbines(OWTs)foundations,whose bearing capacities are the core issue in foundation design.In this paper,the BP neural network and Monte-Carlo simulation coupled method were improved and used to study the reliability of the bearing capac⁃ity of a large-diameter monopile wind turbine under serviceability limit state(SLS),considering the real geo⁃logical condition and the correlation of wind speed,wave height and wave period.The bearing capacity analy⁃sis was conducted based on the monopile-soil contact surface model.The accuracy of the numerical model was verified in two conditions:sand and clay foundations,respectively.Then the finite element model was ad⁃opted to determine the values of the training points of neural network.Finally,with the LW 8MW OWT taken as an example,the reliability of the bearing capacity of the monopile-supported OWT under SLS was calcu⁃lated.The improved method can provide a reference for the subsequent design and construction of offshore wind farms in China.