摘要
针对非线性、非稳态、含噪原始信号混合且混合信号数目小于源信号数目的旋转机械调制故障源信号盲分离问题,提出了一种基于经验模态分解(EMD)和主成分分析(PCA)相结合的方法.对混合信号进行经验模态分解提取嵌入在信号中的所有振荡模式,应用主成分分析方法对所提取的模式进行共性分析,得到模式中的主要成分.利用该方法对仿真数据和两通道滚动轴承加速度振动数据进行了分析,结果表明,该方法能够有效突出旋转机械的故障特征频率成分,避免了误诊断,且适用范围优于独立分量分析方法.
An approach based on empirical mode decomposition （EMD） and principal component analysis （PCA） was presented to deal with the blind source separation （BSS） problem of rotation machines in the case of nonlinear, non stationary, noisy source mixing and the number of observed mixtures being less than that of contributing sources. EMD method was used to extract all oscillatory modes embedded in the observed signals, then PCA method was used to aggregate similar modes into unifying components. The method was applied to analyze the simulation signals and the two-way roller bearing acceleration vibration signals. Analysis result shows that this method can identify the rotation machine＇s fault characteristic frequency clearly and that the application scope is larger than that of independent component analysis （ICA） method.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2006年第2期258-261,共4页
Journal of Zhejiang University：Engineering Science
关键词
经验模态分解
主成分分析
调制
盲源分离
empirical mode decomposition （EMD）
principal component analysis （PCA）
modulation
blind source separation （BSS）
