摘要
提出了一种改进的小波阈值新方法对内燃机缸盖振动信号进行消噪,进而实现特征向量的提取。试验表明采用改进小波阈值方法能有效地消除信号中噪声的干扰,提高信号的信噪比。用小波包提取消噪后信号的能量作为特征向量,来表征内燃机故障特征,可为神经网络的自适应故障诊断提供新的故障样本。
An improved method in wavelet threshold is proposed for the de-noising and feature extraction of vibration signals of engine's cylinder head. Test results show that this method can eliminate the interference of noise effectively and improve the ratio of signal-to-noise. Using wavelet packet to extract the energy of signals after de-noising as the eigenvector could reflect the features of engines' faults, and new fault samples for neural network adaptive fault diagnosis is provided in consequence.
出处
《柴油机》
2007年第5期37-39,53,共4页
Diesel Engine
基金
云南省高校教学科研带头人资助项目
关键词
改进小波阈值消噪
特征提取
内燃机振动
故障诊断
小波包
improved wavelet threshold de-noising
feature extraction
engine vibration
fault diagnosis
wavelet packet