摘要
针对液压泵故障特征提取问题,提出了一种基于奇异值分解和小波包变换的液压泵振动信号特征提取方法。通过奇异值分解将噪声非均匀分布的液压泵振动信号正交分解为噪声分布相对均匀的分量,对各分量进行小波包阈值去噪,重构去噪后分量,对去噪后信号进行小波包分解,提取各频带能量特征。以齿轮泵为例,将该方法对齿轮泵的气穴故障、齿轮磨损和侧板磨损3种常见故障和正常状态的振动信号进行特征提取分析,结果表明,该方法可有效提取齿轮泵故障特征。
A fault feature extraction method for hydraulic pump vibration signal is presented, based on singular value decomposition （SVD） and wavelet packets analysis. In the method, the noise signal is decomposed into the vectors with relatively uniform noise distributions. Then, through wavelet packets threshold de-noising on every vector, the de-noised vectors are even- tually reconstructed . De-noised signal is decomposed by wavelet packets, thus extracting ev- ery power of frequency bands. Taking the gear pump as an example, the features of gear pump air pocket fault, gear abrasion, and side plate abrasion are extracted by using the method, as well as normal state vibration signal. The result indicates that the method can effectively extract the features of gear pump.
出处
《数据采集与处理》
CSCD
北大核心
2012年第2期241-247,共7页
Journal of Data Acquisition and Processing
基金
国防预研基金(9140A27020309JB4701)资助项目
第二炮兵工程学院科技创新基金(XY2010JJB38)资助项目
关键词
液压泵
奇异值分解
特征提取
小波包分析
频带能量
hydraulic pump
singular value decomposition （SVD）
feature extraction
wavelet packets analysis
power of frequency band
