对设备进行故障诊断的主要方法就是测量故障

设备的振动或噪声, 并对其进行分析, 从而找出故障原因。然而振动或噪声信号中除了对分析故障有用的信息外, 还有大量的噪声成分。只有有效地滤除噪声, 才能获得有用的信息, 从而得到可靠的分析结论。传统的滤噪方法是将被噪声污染的信号通过一个滤波器, 滤掉噪声频率成分。但对于短时瞬态信号、非平稳信号、含宽带噪声的信号, 采用传统处理方法有着明显的局限性。小波变换为信号去噪提供了一种有效的方法, 小波阈值去噪具有传统方法不可比拟的优越性。但是小波分解的频域重叠性和阈值选取的不确定性, 使得小波阈值去噪法有也不能得到理想效果。-Equipment fault diagnosis method is to measure the fault

Equipment vibration or noise, and its analysis in order to identify the cause of the malfunction. Vibration or noise signals, however, in addition to the analyze fault useful information, there is a lot of noise components. Only effectively filter out the noise in order to obtain useful information, to obtain reliable conclusions. Traditional filtering method based noise pollution signal through a filter to filter out the noise frequency components. But for the short-term transient signal, non-stationary signals, including broadband noise signal, using the traditional approach has obvious limitations. The wavelet transform provides an effective method for signal de-noising, wavelet thresholding has incomparable superiority of traditional methods. Wavelet frequency domain overlap threshold uncertainty, the wavelet thresholding method can not get the desired effect.