基于S变换的时域边际谱及其应用

为了更好地处理非平稳、非线性振动信号,依据Hilbert-Huang变换边际谱的思想,提出一种基于S变换的时域边际谱,并给出了利用该时域边际谱进行频谱分析的具体方法。对仿真信号的处理结果说明,该方法对振动信号中的冲击能量比较敏感,能够很好地提取冲击的特征频率,同时具有抑制信号高频成分和突出信号低频成分的特点。利用该方法对实际的振动信号进行处理,对滚动轴承故障的成功识别说明该方法能够提取微弱冲击特征频率,体现了其在振动信号频谱分析中的价值。

Marginal Spectrum Based on S Transform and Its Application

In order to deal with non-stationary and nonlinear vibration, this paper proposes a time-domain marginal spectrum based on the S transform, and the marginal spectrum is given by the time domain vibration signal spectrum analysis method. The simulation signal processing show that the method is more sensitive to impact energy of vibration signal; it is able to extract the characteristic frequency; it can inhibit signal high frequency components and highlight the characteristics of the low frequency components. The successful recognition of the rolling bearing fault shows that the method can extract weak impact characteristic frequency, which further embody the value of the vibration signal spectrum analysis.