摘要
针对柴油机缸盖振动信号的特征提取问题,提出了一种基于双树复小波包变换和自适应块阈值降噪的标准化相对能量提取方法,双树复小波包利用并行双树实小波变换分解系数达到信息互补,从而获得近似平移不变性和减少了信息的丢失。自适应分块阈值能够随所分析的信号自适应估计最优阈值,达到更好的降噪效果,同时引入消除频率混叠的算法,抑制了双树复小波包分解过程中虚假频率的产生。仿真信号和试验分析该方法能够更有效地消除噪声影响,所提取的相对能量特征具有更好的可区分度。
In order to extract the relative energy feature of engine cylinder head vibration signals,a new method based on dual-tree wavelet package transformation and adaptive block threshold de-noising was proposed.The dual-tree complex wavelet package transformation(DT-CWPT)made sure that the coefficients decomposed by tree a and tree b could be complementary information at each scale,so the DT-CWPT could achieve approximate shift invariance and decrease the information missing.The adaptive block threshold de-noising method gained the best threshold λ and the optimal neighborhood effect length L for signal de-noising and could obtain higher signal-to-noise ratio.Also the causes of frequency aliasing and frequency bands derangement were discussed and Fourier transformation method was applied to suppress the false frequency in the dual-tree wavelet package transformation.The simulated signal and the measured signal test results showed that the proposed method has a good de-noising performance and it is more effective in fault feature extraction.
出处
《振动与冲击》
EI
CSCD
北大核心
2010年第4期160-163,176,共5页
Journal of Vibration and Shock
关键词
双树复小波包
特征提取
降噪
振动信号
故障诊断
dual-tree wavelet package transformation
feature extraction
de-noising
vibration signal
fault diagnosis
