小波消噪及其在往复泵振动监测信号处理中的应用

WAVELET DE-NOISING AND ITS APPLICATION TO THE PROCESSING OF VIBRATION MONITORING SIGNAL OF RECIPROCATING PUMPS

采用浮动阀值分离噪声是小波分析中一个新的信号检测方法。本文在简述了小波分析及噪声分离的非线性小波理论的基础上 ,给出了从噪声污染信号中恢复原信号的实例。针对三缸往复泵测量信号非平稳时变特点 ,研究了用小波消噪方法提取反映往复泵动力端状态的振动波形特征。通过对振动信号的波形特征提取表明 。

De_noisding by soft_thresholding is a new signal detection method in wavelet analysis.On the base of briefly introducing the wavelet analysis theory and nonlinear wavelet de_noising theory by soft_thresholding,some examples of signal recovery with white_noise heavily polluted are given in the paper.With a view to time_varying characteristic of the vibration signal measured on triplex reciprocating pump,an approach on the waveform feature extraction which reflects the various conditions of reciprocating pump with nonlinear wavelet de_noising analysis method is presented.The study shows that with the method offered the useful information reflecting the condition of the reciprocating pump can be effectively extracted from the highly noised signal by soft_thresholding de_noising method at the power end of the reciprocating pump.