摘要
提出了一种基于信号特征的自适应提升小波方法,即以提升小波为基础,根据信号分解后的熵来选择预测滤波器系数和更新滤波器系数,它克服了传统小波变换的不足,和提升小波只能依据信号特征来设计预测滤波器,而不能设计更新滤波器的问题。该方法用于往复机械气阀的振动信号特征提取,有效地提取了气阀的故障特征信号。实验中采用不同的小波对信号进行降噪性能比较,自适应提升方法设计的小波明显优于实验室中采用的其它小波。
Adaptive Lifting wavelet based on signal features is presented, and the entropy after signal decomposition is adopted to select predict filter coefficient and update filter coefficient. This method has avoided the shortcomings in the traditional wavelet transformation and the problems that the lifting wavelet can only design the predict filter according to signal feature, instead of update filter. It is introduced in the feature extraction of the vibration signal induced from reciprocating machinery valve, and the fault characteristic signal can be extracted effectively. According to signal denoising performance comparison using different wavelet in the experiment, wavelet based on adaptive lifting method is obviously superior to others.
出处
《流体机械》
CSCD
北大核心
2007年第4期23-27,共5页
Fluid Machinery
基金
国家自然科学基金项目(50375014
50575016)
关键词
自适应提升
往复机械
小波
特征提取
adaptive Lifting
reciprocating machinery
wavelet
feature extraction
