一种确定奇异值分解降噪有效秩阶次的改进方法

PDF(2088 KB)

振动与冲击 ›› 2014, Vol. 33 ›› Issue (12) : 176-180.

论文

一种确定奇异值分解降噪有效秩阶次的改进方法

王建国，李健，刘颖源

作者信息 +

Method for determining the effective rank of SVD denoising

WANG Jian-guo，LI Jian，LIU Ying-yuan

Author information +

文章历史 +

摘要

提出了一种基于奇异值差分谱单边极大值的降噪方法，实现了对旋转设备故障信号信噪比的提高。首先通过相空间重构Hankel矩阵的方法对原始振动信号进行处理，再进行奇异值分解，最后采用奇异值差分谱单边极大值原则确定所含的较大峰值降噪阶数。通过数值仿真和实际轴承故障数据的应用分析，表明了该方法可以有效地提高信号的信噪比，为后期的故障特征信号提取创造有利条件。

Abstract

A new effective method is presented to solve denoising based on difference spectrum unilateral maxima of singular value.In order to improve the signal-to-noise ratio of the rotating equipment fault signal. In order to improve the signal-to-noise ratio of the rotating equipment fault signal. Firstly, the phase space reconstruction Hankel matrix of the original vibration signal, and then the trajectory attractor matrix was decomposed by using Singular Value Decomposition. Finally，according to the difference spectrum unilateral maxima principle of singular value to determine the reconstruction order of singular value．Though numerical simulation and the practical application of the bearing failure data analysis. The results show that the method can effectively improve the signal-to-noise ratio of the signal, and create obvious advantage for late fault characteristic signal extraction.

导出引用

王建国;李健;刘颖源. 一种确定奇异值分解降噪有效秩阶次的改进方法[J]. 振动与冲击, 2014, 33(12): 176-180

WANG Jian-guo;LI Jian;LIU Ying-yuan. Method for determining the effective rank of SVD denoising[J]. Journal of Vibration and Shock, 2014, 33(12): 176-180