摘要
为提高一维振动信号的数据压缩比,提出了基于小波域奇异值分解的信号压缩方法。首先,将工程采集的振动信号进行小波分解,通过补零对不同尺度的小波系数构建矩阵;然后,对小波系数矩阵进行奇异值分解,根据奇异值累积贡献度确定奇异值及其对应的左、右奇异向量,实现信号的压缩。将该方法应用于轴承故障信号,大大减少了数据量,取得了良好的效果。
In order to improve the data compression ratio of one-dimensional vibration signal,a signal compression method is proposed based on singular value decomposition in wavelet domain.Firstly,a vibration signal is sampled and decomposed by wavelet.Then a wavelet coefficient matrix is built by zero filling and analyzed by singular value decomposition.The singular values and their corresponding left and right singular value vectors are obtained according to singnlar value accumulation contribution rate,the signal compression is realized.The method is applied to bearing fault signal,and the data volume is greatly reduced and a good result is got.
出处
《轴承》
北大核心
2013年第11期51-54,共4页
Bearing
关键词
滚动轴承
小波分解
奇异值分解
数据压缩
rolling bearing
wavelet decomposition
singular value decomposition（SVD）
data compression
