摘要
超声回波信号反映了润滑油中磨粒的大量信息。为了提取淹没在强噪声环境下的超声回波信号,提出了一种基于双树复小波变换(DT-CWT)的油液磨粒超声散射回波信号去噪新方法。利用双树复小波变换具有近似平移不变性和有效去噪等优点,首先对超声散射回波信号进行双树复小波分解,然后对分解得到的高频系数进行阈值处理,最后进行双树复小波重构。结果表明:分解层数为6层时,去噪后信号的信噪比更高、均方误差更小、相似系数更大、幅值最大偏差更小。双树复小波变换硬阈值去噪效果比传统小波去噪效果明显好。
The ultrasonic echo signal reflecting the debris lubricant contains a lots of information. In order to extract the ultrasonic echo signal submerged in strong background noise,a new de-noising method of ultrasonic scattering echo signal for oil wear debris based on the Dual-Tree Complex Wavelet Transform( DT-CWT) is proposed. Firstly,for the DT-CWT having the approximate shift-invariant and effective de-noising,the ultrasonic scattering echo signals are decomposed by using the DT-CWT,and then the threshold processing of high frequency coefficients is performed,finally,the signal is reconstructed by using the DT-CWT. The simulated and experimental results show that the SNR of the de-noising signal is higher,the RMSE is smaller,the NCC is higher,and the maximum amplitude difference( MAD) is smaller at a decomposition order of six. The DT-CWT de-noising method by using a hard threshold is more obvious than the traditional wavelet de-noising method.
出处
《机械科学与技术》
CSCD
北大核心
2015年第2期229-233,共5页
Mechanical Science and Technology for Aerospace Engineering
基金
国家自然科学基金项目(50705097
51305454)
清华大学摩擦学国家重点实验室开放基金项目(SKLTKF09B06)资助
关键词
双树复小波变换
超声回波信号
分解层数
去噪
油液磨粒
computer simulation
decomposition order
design of experiments
de-noising
DT-CWT
efficiency
experiments
lubricants
mean square error
noise abatement
oil wear debris
schematic diagrams
signal reconstruction
signal to noise ratio
ultrasonic echo signal
ultrasonics
wavelet transforms
