|
Ji Zhong
Jin Tao
Qin Shuren
College of Mechanical Engineering,
Chongqing University,
Chongqing 400030, China |
|
|
SIGNAL FEATURE
EXTRACTION
BASED UPON
INDEPENDENT COMPONENT
ANALYSIS AND
WAVELET TRANSFORM*
Abstract: It is an important precondition for machine fault diagnosis that vibration signal can be extracted effectively. Based on the characteristic of noise interfused during the course of sampling vibration signal, independent component analysis (ICA) method is combined with wavelet to de-noise. Firstly, The sampled signal can be separated with ICA, then the function of frequency band chosen with multi-resolution wavelet transform can be used to judge whether the stochastic disturbance singular signal is interfused. By these ways, the vibration signals can be extracted effectively, which provides favorable condition for subsequent feature detection of vibration signal and fault diagnosis.
Key words:
Independent component analysis (ICA) Wavelet transform De-noising Fault diagnosis Feature extraction
|