基于惯性网络的兆瓦级风电叶片形变测量算法研究  

RESEARCH ON DEFORMATION MEASUREMENT ALGORITHM OF MEGAWATT WIND TURBINE BLADE BASED ON INERTIAL NETWORK

作　　者：张华强[1] 刘林 秦昌礼 刘卫生 苏庆华[3] 陈雨 Zhang Huaqiang;Liu Lin;Qin Changli;Liu Weisheng;Su Qinghua;Chen Yu(School of Mechanical Engineering,Shandong University of Technology,Zibo 255049,China;Lianyungang Zhongfu Lianzhong Composite Group Company Limited,Lianyungang 222000,China;School of Information Engineering,Beijing Wuzi University,Beijing 101149,China;Military Representative Office of Rocket Army in 211 Factory,Beijing 100076,China)

机构地区：[1]山东理工大学机械工程学院,淄博255049 [2]中复连众复合材料集团有限公司,连云港222000 [3]北京物资学院信息学院,北京101149 [4]北京航天发射技术研究所,北京100076

出　　处：《太阳能学报》2022年第10期186-191,共6页Acta Energiae Solaris Sinica

基　　金：国家自然科学基金青年基金(61803035)。

摘　　要：针对目前风电叶片形变过程中挠曲度测量误差较大的问题,提出一种基于惯性网络的相对运动监测算法。首先根据应力分析对叶片建立主从式惯性网络,然后推导出主子节点间的相对运动解算方法并建立相对导航误差模型,设计相对导航误差估计滤波器,通过误差反馈保证子节点的位姿解算精度;其次构建联邦式的相对惯性导航误差估计滤波器进行主子节点的数据融合,提高风电叶片的形变估计精度及系统整体的容错性;最后对某型风电叶片进行静力加载试验,试验结果表明:该算法可准确测量出叶片上各节点在三维空间的位移曲线,3个子节点在挥舞方向的平均相对误差仅为0.92%、1.18%、1.07%。该算法可实时监测风电叶片的位姿状态,在叶片检测及风力机日常运行的安全监测上具有较好的理论研究意义与工程应用价值。To deal with the large flexure measurement error in the deformation process of wind turbine blade,a relative motion monitoring algorithm based on inertial network is proposed.Firstly,the master-slave inertial network is established according to the stress analysis of wind turbine blade,then the calculating method for the movement of the master IMU relative to the slave IMU is derived and the relative navigation error model is established.The relative navigation error estimation filter is designed to ensure the accuracy of the posture solution of each slave nodes through the error feedback.Secondly,a federated relative inertial navigation error estimation filter is constructed for data fusion of master node and slave nodes,which improves the deformation estimation accuracy of the integral wind turbine blade and the fault tolerance of the entire system.Finally,a static loading test is carried out with a certain type of wind turbine blade.The test results show that algorithm can accurately measure the displacement curve of each node on the wind turbine blade,the mean relative errors(MREs)of the three slave nodes in wave direction are 0.92%,1.18%and 1.07%,respectively.This algorithm can monitor the position and attitude of wind turbine blade in real time,and it has good theoretical research significance and engineering application value in wind turbine blade detection and wind turbine safety monitoring.

关 键 词：风电叶片 形变 惯性网络 相对运动 数据融合 

分 类 号：TH137[机械工程—机械制造及自动化]