一种转子启停车过程的基频振动分量提取方法
New Signal Processing for Extracting Fundamental Vibration Component of Rotor in Startup or Shut Down Process

针对传统信号处理算法在分离转子启停车过程中包含复杂非平稳谐波分量的振动响应信号时存在计算量大、对宽频信号提取能力差等问题,基于零相移滤波和相位解调原理,提出了一种转子启停车过程中的基频振动分量提取方法。该方法通过一次非线性相位解调计算,把启停车信号各分量在频带上进行重新分布,这样宽频的基频分量被解调到可见的窄带上,仅在频域即可实现基频分量与其他高次谐波分量的分离,再通过零相移滤波即可提取这个窄带分量。仿真计算和实验数据分析结果表明:与传统的信号处理算法在处理转子启停车信号上相比,提出的方法具有原理简单易实现、计算速度快、受相邻分量带宽限制小、滤波器参数能够从频谱上直接选取等优势,同时该方法提取出的基频振动分量波形平整,无相位偏移,满足进一步的分析计算,为转子启停车信号的快速提取提供了一种新的方式。
For the traditional signal processing algorithms, it is difficult to process the complicated vibration signals of rotor in startup or shut down process due to the huge calculation task and poor ability to extract wide？？band signals. Following zero？？phase filter and phase demodulation principle, a novel method for extracting the fundamental vibration component of rotor is proposed. The wide？？band fundamental vibration component can be demodulated into a narrow？？band interval via once nonlinear phase demodulation, which is segregated with the band of other higher？？order harmonics in frequency domain, then the fundamental component is extracted by a zero？？phase filter. The simulation calculation and experimental validation demonstrate that the proposed method outperforms the traditional signal processing algorithms by its simple mechanism, quick calculation and little limit to the band between adjacent components. The visual selection of filtering parameters can be directly realized in frequency domain and the extracted amplitude and phase of the fundamental components are accurate for the further calculation and analysis