华南理工大学学报(自然科学版) ›› 2016, Vol. 44 ›› Issue (10): 1-7.doi: 10.3969/j.issn.1000-565X.2016.10.001

• 机械工程 •    下一篇

基于谐波小波和SVD 的可倾瓦滑动轴承工频干扰抑制

张晓涛 李伟光   

  1. 华南理工大学 机械与汽车工程学院,广东 广州 510640
  • 收稿日期:2016-01-11 修回日期:2016-05-10 出版日期:2016-10-25 发布日期:2016-09-01
  • 通信作者: 张晓涛( 1973-) ,男,博士生,主要从事旋转机械故障诊断研究. E-mail:zhang422429@163.com
  • 作者简介:张晓涛( 1973-) ,男,博士生,主要从事旋转机械故障诊断研究.
  • 基金资助:

    国家高技术研究发展计划( 863 计划) 项目( 2015AA043005) ; 南沙区科技计划项目( 2014CX07)

Power Interference Removal of Tilting-Pad Journal Bearing Based on Harmonic Wavelet and SVD

ZHANG Xiao-tao LI Wei-guang   

  1. School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China
  • Received:2016-01-11 Revised:2016-05-10 Online:2016-10-25 Published:2016-09-01
  • Contact: 张晓涛( 1973-) ,男,博士生,主要从事旋转机械故障诊断研究. E-mail:zhang422429@163.com
  • About author:张晓涛( 1973-) ,男,博士生,主要从事旋转机械故障诊断研究.
  • Supported by:
    Supported by the National High-tech R&D Program of China( 863 Program) ( 2015AA043005)

摘要: 针对新型流体支撑可倾瓦滑动轴承的振动位移信号中的工频干扰问题,提出了基于谐波小波和奇异值分解( SVD) 的工频干扰抑制算法. 首先基于谐波小波的严格盒型频谱和任意频段任意细分的特性,提取包含工频频谱的频段,然后在时域构建Hankel 矩阵并进行奇异值分解,将反映工频成分的特征值置零,从而抑制工频成分,最后将去除工频的子带频谱同原始频谱其他子带合并,得到去除工频干扰的有用信号用于后续分析. 试验结果表明,此算法可有效抑制工频干扰,为后续的新型可倾瓦滑动轴承的振动特性研究奠定基础,具有一定的工程应用价值.

关键词: 谐波小波, 奇异值分解, 工频干扰, 可倾瓦滑动轴承

Abstract:

In order to remove the power interference in the vibration displacement signals of a novel tilting-pad journal bearing with fluid pivot,a power interference removal algorithm based on the harmonic wavelet and the singular value decomposition ( SVD) is proposed.In the algorithm,first,the frequency domain containing the power frequency is extracted by using the harmonic wavelet that has strict box-like spectrum and arbitrary detail of arbitrary frequency domain.Then,a Hankel matrix is constructed in the time domain,the singular value decomposition is performed,and the singular values relevant to the power frequency are set to be zero,thus removing the power frequency.Finally,all the sub-bands without power frequency are merged with the other sub-bands of the original frequency spectra.Thus,the time domain signals without the power interference are achieved for a further study.Experiment results show that the proposed algorithm can effectively remove the power interference.This algorithm helps to investigate the vibration characteristic of the novel tilting-pad journal bearing,and it is both practical and feasible.

Key words: harmonic wavelet, singular value decomposition, power interference, tilting-pad journal bearing