高速旋转机械用消除频率混叠现象的小波变换算法研究

高速旋转机械由于其工作转速高、控制精度高，其故障信号又属于微弱突变信号，造成常规监测方法准确率偏低，成为制约高速旋转机械故障监测进一步发展的最大障碍。本文设计了一种消除小波分解和重构过程中产生的频率混叠的算法。该算法利用傅里叶变换和傅里叶逆变换构成Mallat小波变换在分解和重构过程中所需的严格正交镜像滤波器，从而达到Mallat小波变换过程中必须具有的理想截止特性，最终去除掉多余的频率成分。利用改进前后的小波算法对实际信号进行分析，结果表明，本文设计的Mallat小波变换改进算法可消除信号分析中出现的频率混叠现象。

Improvement Algorithm of Wavelet Transformation of High Speed Rotating Machine for Reducing Phenomenon of Frequency Mixed

Due to high speed and high precision, the accuracy of detecting method of high speed rotating machine is low, and low accuracy is the biggest obstacle to status detection. A method of reducing frequency mixed in the procedure between resolvement and reconstitution of wavelet was designed in this paper. The algorithm uses Fourier transformation and inverse Fourier transformation to construct orthogonal image filter bank in the procedure of resolvement and reconstitution, so as to achieve ideal cut-off-eigenschaft in Mallat wavelet transformation, and at the end, redundant frequency can be reduced. The wavelet algorithms before and after improvement were used to analyze the actual signal. The results show that the improvement Mallat wavelet algorithm can reduce frequency mixed in the signal analysis.