DocumentCode :
599040
Title :
Feature extraction method of mechanical impulse based on nonlinear manifold learning
Author :
Lin Liang ; Maolin Li ; Guanghua Xu ; Huizhong Gao
Author_Institution :
Sch. of Mech. Eng., Xi´´an Jiaotong Univ., Xi´´an, China
fYear :
2012
fDate :
16-18 Oct. 2012
Firstpage :
1806
Lastpage :
1810
Abstract :
Feature extraction is of great importance in condition monitoring of machinery. Manifold learning theories brought a new idea for recognizing and predicting the underlying nonlinear behavior. In this paper, in order to extract the key feature of mechanical signal, a principle manifold feature extraction method based on the local tangent space alignment is proposed. Integrated with reconstruction of phase space, the method can extract the manifold feature which provides a more truthful low dimensional representation. During the searching of embedding manifold, integrated with the advantage of the kurtosis and skewness indexes, the selection of local neighborhood parameters is introduced to evaluate the feature. The industrial measurements show that this approach, compared with the wavelet soft-threshold and the orthogonal matching pursuit methods, is more effective to extract the weak periodic features from mechanical signals.
Keywords :
condition monitoring; feature extraction; learning (artificial intelligence); machinery; mechanical engineering computing; signal reconstruction; industrial measurements; kurtosis indexes; local neighborhood parameter selection; local tangent space alignment; machinery condition monitoring; mechanical impulse; mechanical signal; nonlinear behavior prediction; nonlinear behavior recognition; nonlinear manifold learning theories; phase space reconstruction; principle manifold feature extraction method; skewness indexes; weak periodic feature extraction; Educational institutions; Entropy; Feature extraction; Indexes; Manifolds; Matching pursuit algorithms; Vibrations; feature extraction; manifold learning; mechanical diagnosis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image and Signal Processing (CISP), 2012 5th International Congress on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4673-0965-3
Type :
conf
DOI :
10.1109/CISP.2012.6470018
Filename :
6470018
Link To Document :
بازگشت