应用经典极值理论对风机极端载荷的预报  

Application of classical extreme value theory to the prediction of extreme load of wind turbines

作　　者：周帅 王迎光[1,2,3] 李昕雪 ZHOU Shuai;WANG Ying-guang;LI Xin-xue(Shanghai Jiaotong University,School of Naval Architecture,Ocean and Civil Engineering,Shanghai 200240,China;State Key Laboratory of Ocean Engineering,Shanghai 200240,China;Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration,Shanghai 200240,China)

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200240 [2]海洋工程国家重点实验室,上海200240 [3]高新船舶与深海开发装备协同创新中心,上海200240

出　　处：《舰船科学技术》2018年第10期93-98,共6页Ship Science and Technology

基　　金：海洋工程国家重点实验室自主研究课题资助项目(GKZD010038)

摘　　要：在风机极端载荷预报中,短期极值分布的拟合效果决定了长期极值预报的准确度。为了解决同类研究中遇到的短期极值分布问题,减少极端载荷预报的误差,引入经典极值理论对外推方法进行研究。以分块法选取样本点,广义极值分布进行拟合,线性矩法估计分布参数,求解了WP_Baseline 1.5 MW陆上风机叶片根部面外弯矩的长期分布,获得了相应的超越概率曲线。着重以1年和20年重现周期下的极端载荷与同类研究中的"真实数据"及外推结果进行对比。与此同时,为了检验数据的稳定性,从仿真时长和分块容量两方面对外推方法进行检验。结果表明,本文所述方法不仅具有较好的精准度,同时具有较高的可靠性。因此为风机极端载荷的预报提供了一种参考。In the extreme load prediction of wind turbines, the fitting effect of the short-term extreme load distribution determines the accuracy of the long-term extreme load. In order to solve the fitting problem of short-term extreme load distribution encountered in similar studies and reduce prediction error, the classical extreme value theory is introduced to find a good extrapolation method. Taking the WP_Baseline 1.5 MW onshore wind turbine as an example, the sample points were selected by block method and the generalized extreme value distribution was fitted and the linear moment method was used to estimate the distribution parameters. Finally, the long-term distribution of the out-of-plane bending moment of the blade root was solved, and the corresponding exceedance probability curve was obtained. The comparison is focused on the extreme load with 1 year and 20 year return periods between the real data and other results in similar studies. At the same time,in order to varify the stability of prediction, the extrapolation method is tested from both simulation time and block capacity.The research shows that the method described in this paper not only has good precision, but also high reliability. So it provides a reference for the extreme load predicting of wind turbines.

关 键 词：风机 极端载荷 重现周期 超越概率 极值理论 

分 类 号：TM315[电气工程—电机]