齿轮故障振动信号检波解调分析中的混频效应

在齿轮故障诊断中，振动信号解调分析是一种很有效的方法。本文在分析调幅调相信号检波解调分析原理的基础上，讨论了该类算法软件实现过程中产生的混频效应，指出信号在检波解调过程中经过非线性变换后将包含高频成分，有可能发生明显的混频，导致得出错误的解调信号。通过合理地选择采样频率及分析算法中低通滤波器的截止频率或采用较高的采样频率能有效地消除这种混频效应对解调的影响，最后以仿真数据及实验论证了这些结论。

Aliasing Effect on Rectification Based on Demodulation Analysis for Diagnosing Gear Fault

Demodulation analysis is widely used for diagnosing gear fault. In this paper, the principle of rectification based demodulation analysis for amplitude and phase modulated signals is analyzed, and the aliasing effect caused by rectification is pointed out and discussed. It is noticed that the rectification operation is nonlinear and gives rise to high frequency components in the discrete signal, which may cause aliasing effect, so additional masking signals may be contained in the demodulated signal. The conditions for the masking signals not appearing are deduced, and it is shown that aliasing effect will have no influence on demodulated signal if the sampling frequency and the parameter used in demodulation are selected appropriately or the sampling frequcney is high enough. These conclusions are verified by both synthetic and experimental data.