特征模态函数双谱分析在叶片裂纹识别中的应用.pdfVIP

第36 卷 第1期 噪 声 与 振 动 控 制 Vol 36 No. 1 2016 年2 月 NOISE AND VIBRATION CONTROL Feb. 2016 文章编号：1006-1355(2016)01-0153-04 特征模态函数双谱分析在叶片裂纹识别中的应用 靳子洋，陆永耕，张 彬，姚晓龙 （上海电机学院 电气学院，上海 200240 ） 摘 要：针对叶片裂纹故障振动信号特征，提出特征模态函数的双谱分析法，首先利用经验模态分解（Empirical Mode Decomposition ，EMD ）对振动信号进行自适应滤波分解，产生一系列不同时间尺度的特征模态函数（Intrinsic Mode Function ，IMF），然后对含有高频信号的高阶IMF 分量进行重构，利用双谱提取叶片裂纹的振动信号特征。通过 仿真信号和实验分析，验证叶片裂纹产生的高频冲击对叶片振动信号高频部分双谱的影响，证明IMF 分量双谱分析的 有效性，为风电叶片正常状态监测提供依据。 关键词：振动与波；叶片裂纹，特征模态函数，经验模态分解，双谱分析 + 中图分类号：TP206 .3 ；TN911.7 文献标识码：A DOI 编码：10.3969/j.issn. 1006-1335.2016.01.033 Application of Bispectrum Analysis of Intrinsic Model Function in the Blade Crack Recognition JIN Zi-yang , LU Yong-geng , ZHANG Bin , YAO Xiao-long ( School of Electric Engineering, Shanghai Dianji University, Shanghai 200240, China ) Abstract : According to the vibration signal characteristics of the blade crack fault, the bispectrum analysis of intrinsic model functions was proposed. Firstly, a series of different time scale characteristics of the intrinsic model functions was produced by the method of Empirical Mode Decomposition (EMD). Then, the high order intrinsic model functions including the high frequency components of the vibration signal were reconstructed, and the bispectrum was used to analyze the feature of the vibration signal. Through the signal simulation and experimental analysis, the changing of the feature of blade vibration signal in the high frequency range was identified with the proposed method. This work is of significance for wind- power blades health co