机电系统非平稳振动信号分析方法的研究

Study on the analytic methods of non-stationary vibration signals of mechanical and electrical system

为了在非平稳状态下的机电系统故障诊断及趋势预测中合理地选择信号分析方法,对短时傅立叶变换、小波分析和希尔波特-黄变换这3种分析非平稳信号较为有效的方法进行了研究。通过对一个仿真典型信号的分析结果进行比较,验证了3种方法的有效性和各自特点,并分析了产生差异的原因,探讨了它们各自适合的应用场合。

In order to select a proper method of the signal's analysis in the fault diagnosis and trend prediction of mechanical and electrical system under non-stationary states, three methods, effective in the analysis of non-stationary signals, are studied. They are short-time Fourier transformation, Wavelet Analysis and Hilbert-Huang transformation. Through the comparison of the analytic results of a simulated typical signal, the validity and characteristic of each method are confirmed, and the causes of differences are analyzed. Furthermore, the suitable application situation for each method are discussed.