摘要
在对海上风机巡检过程中,由于无人机螺旋桨振力振动信号数据量较大,且海上环境的复杂性和不稳定性增加了异常信号的干扰,使得异常信号的检测和识别更加困难。为此,该文提出一种海上风机无人机巡检螺旋桨振力异常监测方法。利用集成经验模态(EEMD)分解,处理无人机螺旋桨振力信号,提取固有模态分量(IMF);利用稀疏傅里叶变换,处理具有周期性成分的IMF组分,提取振动强度异常特征;引入多传感器信息决策融合网络结构,并利用神经网络对其展开融合和综合决策,最终实现无人机螺旋桨振力异常监测。实验结果表明,所提方法可以获取高准确率和高效率的螺旋桨振力异常监测结果。
During the inspection process of offshore wind turbines,due to the large amount of data on the vibration signals of unmanned aerial vehicle propellers and the complexity and instability of the marine environment,the interference of abnormal signals is increased,making the detection and identification of abnormal signals more difficult.To this end,a monitoring method for abnormal propeller vibration force during offshore wind turbine unmanned aerial vehicle inspection is proposed.Utilizing integrated empirical mode decomposition(EEMD)decomposition to process the vibration signal of unmanned aerial vehicle propeller and extract the intrinsic mode function(IMF)components.Using sparse Fourier transform to process IMF components with periodic components and extract abnormal vibration intensity features.Introducing a multi-sensor information fusion network structure and utilizing neural networks to fuse and make comprehensive decisions,ultimately achieving abnormal monitoring of unmanned aerial vehicle propeller vibration force.The experimental results show that the proposed method can obtain high accuracy and efficiency monitoring results of abnormal propeller vibration force.
作者
刘艳贵
傅望安
王海明
曾崇济
LIU Yangui;FU Wang'an;WANG Haiming;ZENG Chongji(Clean Energy Branch of Huaneng(Zhejiang)Energy Development Co.,Ltd.,Hangzhou 310014,China;China Huaneng Clean Energy Research Institute,Beijing 102209,China)
出处
《自动化与仪表》
2024年第1期93-97,共5页
Automation & Instrumentation
关键词
海上风机无人机
巡检螺旋桨
振力异常
监测
offshore wind turbine unmanned aerial vehicle
patrol inspection of propellers
abnormal vibration force
monitor
