Feature Selection Algorithm Based on EMD and GA-PLS and its Application
Li Sheng ; Zhang Pei-lin
Author information+
{{custom_zuoZheDiZhi}}
{{custom_authorNodes}}
{{custom_bio.content}}
{{custom_bio.content}}
{{custom_authorNodes}}
Show less
History+
Received
Revised
Published
2010-11-23
2011-01-13
2012-02-28
Issue Date
2012-02-28
Abstract
In order to solve problems of the nonstationarity of vibration signal and the optimization of feature selection, feature selection algorithm based on EMD and GA-PLS is proposed. In this algorithm, firstly, EMD method is used to decompose the vibration signals into a number of intrinsic mode function components, and the auto-regressive mode (AR) model of each IMF component is established. The main auto-regressive parameters and the loss function are regarded as original feature vectors. Then, genetic algorithm-partial least squares (GA-PLS) algorithm is used to selecting new feature vectors, which are high correlation with fault information, from original feature vectors. Finally, when these new feature vectors are used as inputs, classifiers are established for identifying the conditions and fault patterns of manual operated directional valve. Experimental results show that all the feature vectors are selected correctly, and the proposed algorithm can be used in the fault diagnosis.
Li Sheng;Zhang Pei-lin.
Feature Selection Algorithm Based on EMD and GA-PLS and its Application[J]. Journal of Vibration and Shock, 2012, 31(4): 134-138
PDF(1657 KB)
