Feature Selection Method for Manual Operated Directional Valve Based on EMD and GA-PLS

Abstract:

A feature selection method for manual operated directional valve based on EMD and GA-PLS is proposed. Vibration signal from a manual operated directional valve is difficult to find the main fault features. The EMD method is used to decompose the vibration signal into a number of intrinsic mode function (IMF) components. Auto-regressive (AR) model is a time-series tool which can represent the important information of system state. Therefore, AR model of each IMF component is established. The AR parameters and the loss function are regarded as the feature vectors. In order to reduce feature vectors, Genetic algorithm-partial least squares (GA-PLS) is a suitable method for selecting effective features. GA-PLS method is used for selecting new features from these feature vectors, which constitute the AR parameters and the loss function. Experimental results show that this method can be applied to select features effectively.

• Periodical: Applied Mechanics and Materials (Volumes 170-173)
• Pages: 2712-2717
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.170-173.2712
• Online since: May 2012
• Authors: Jing Li Qi, Sun Fu, Jian Si Chen
• Keywords: Auto-Regressive Model, Empirical Mode Decomposition (EMD), Feature Selection, Genetic Algorithm-Partial Least Squares, Manual Operated Directional Valve

References

[1] N.E. Huang, Z.Shen, S.R. Long, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceeding of the Royal Society A, 1998, 454(1971): 903-995.

DOI: 10.1098/rspa.1998.0193

[2] N.E. Huang, Z.Shen, S.R. Long. A new view of nonlinear water waves: the Hilbert spectrum.. Annual Review of Fluid Mechanics, 1999, 31: 417-457.

DOI: 10.1146/annurev.fluid.31.1.417

[3] LI Ting, JIANG Qing, CAI Jinhu, et al. Four-way Valve Fault Diagnosis Based on Wavelet Denosing and HHT. Journal of China University of Metrology, 2009，20(3):235-239.

[4] Yu Dejie, Chen Miaofeng, Cheng Junsheng. A Fault Diagnosis Approach for Rotor Systems Based on Empirical Mode Decomposition Method and Support Vector Machines. Proceedings of the CSEE, 2006, 26(16): 162-167.

[5] Jie Peizhong, Zhang Zhiyong. New Appraoch to Conform Feature Samples for Faults Diagnosis Classifiers. Systems Engineering and Electronic.2001，23(11) 1589-1592.

[6] CHENG Junsheng, YU Dejie, TANG Jiashi and et al. Application of Correlation Dimension and EMD-Based AR model in the Fault Diagnosis. Systems Engineering and Electronic.2007，29(9) 1589-1592.

[7] Gonzalez A L. Leardi R. Genetic algorithms applied to feature selection in PLS regression:how and when to use them. Chemometr Intell Lab.1998，41(2):195-208.

DOI: 10.1016/s0169-7439(98)00051-3

[8] Leardi R. Application of genetic algorithm-PLS for feature selection in spectral data sets. J Chemometr.2000，14(5/6):643.

DOI: 10.1002/1099-128x(200009/12)14:5/6<643::aid-cem621>3.0.co;2-e