基于第2代小波的水电机组振动信号预处理

Vibration Signal Preprocessing of Hydraulic Generator Set Based on the Second Generation Wavelet

传统的小波去噪效果与选用的小波基函数有关,小波基函数选择不当会把信号的有用成分当成噪声去掉,给故障诊断造成困难。而基于提升算法的第2代小波变换可以通过设计预测系数和提升系数获得具有某种特性的小波基函数。文中提出采用基于提升算法的第2代小波变换来进行水电机组振动信号的去噪处理,通过分析提升算法的基本原理,使用第2代小波对水电机组振动信号在不同分辨率下进行分解处理,并使用阈值处理小波分解的细节系数,将处理过的细节系数和近似系数一起重构信号。仿真结果表明,采用第2代小波变换要优于传统小波变换的去噪效果,第2代小波在信号处理中有广阔的应用前景。

As the de-noise result of traditional wavelet is related the wavelet basis function, the fault diagnosis would be difficulty while the wavelet basis function isn＇t suitable for the vibrant signal. In the second generation wavelet transform （SGWT）, wavelet basis function with some special characteristic can be obtained by means of designing prediction and update coefficient. SGWT based on the lifting scheme in de-noising vibrant signal processing of hydraulic generator set was proposed. Through analyzing the principle of lifting scheme, it is fetched out that it de-composes the vibrant signal at the different resolution, and process the detail coefficient of wavelet transform by threshold of limiting range. The processed detail coefficient and the approximate coefficient are used to construct the vibrant signal. The simulation result shows that the second generation wavelet transform is more excellent, contrasted with the traditional wavelet transform, and it has a wide foreground.