复局部均值分解全矢包络技术及其在转子故障特征提取中的应用

Full Vector Envelope Technique Based on Complex Local Mean Decomposition and Its Application in Fault Feature Extraction for Rotor System

为更全面提取转子故障特征,将全矢谱和复局部均值分解(Complex local mean decomposition,CLMD)相结合,提出二元的全矢包络技术——CLMD全矢包络技术。采用正交采样技术获取转子同一截面上互相垂直方向上的振动信号,并将其组成一个复数信号;运用CLMD将复数信号按能量从高到低的顺序依次分离出系列复乘积函数(Complex product function,CPF),并解调出CPF的复包络;由于故障特征主要在能量较高的CPF分量中,通过全矢谱技术融合前几阶CPF分量的包络信号,得到相应的全矢包络谱。仿真的调幅-调频信号分析结果表面,较之Hilbert解调,CLMD全矢包络技术可提取隐含的调频信息,而且不存在虚假的低频谱线。转子试验台模拟的基座松动信号、碰摩信号分析结果表明,较之单源信息的包络谱,CLMD全矢包络技术提取的谱线特征更清晰、全面,而且根据全矢包络谱可有效区分基座松动引起的碰摩和单一碰摩故障。

A bivariate full vector envelope technique is proposed based on complex local mean decomposition（CLMD）, which combines full vector spectrum（FVS） and CLMD, to extract comprehensive rotor fault feature. Orthogonal sampling technique is used to obtain the same section of the rotor vibration signal in the direction perpendicular to each other and composited them to a complex signal. Complex signal is divided into series of complex product functions（CPF） followed in descending order by energy and the complex envelope signal of CPF is demodulated. Since the fault characteristics are mainly concentrated in the higher-energy component of the CPF, complex envelope signal of the first few CPF component by FVS is fused to get corresponding full vector envelope spectrum. Simulation AM- FM signal analysis results show that, compared with Hilbert demodulation, CLMD full vector envelope technology extracts information of implied FM, but there is no false low-frequency spectrum. Analysis of the signal of base loose and rubbing rotor from rotor test rig show that, compared with single-source information envelope spectrum, spectral features extracting by CLMD full vector envelope technology are not only more clear and comprehensive, also distinguish rub caused by base loose with the single rub using CLMD full vector envelope technology.