Rolling bearing fault diagnosis based on local wave method
and KPCA-LSSVM
YANG Xian-yong1,2, ZHOU Xiao-jun1, ZHANG Wen-bin1, YANG Fu-chun1
1. Zhejiang Provincial Key Laboratory of Advanced Manufacturing Technology, Zhejiang University,
Hangzhou 310027, China; 2. China Ship Development and Design Center, Wuhan 430064, China
Aimed at the nonstationary characteristics of rolling bearing vibration signal, a fault diagnosis method was proposed based on localwave method and KPCA(kernel principal component analysis) LSSVM(least squares support vector machine). Firstly, local wave decomposition was used to decompose rolling bearing vibration signal into several intrinsic mode function (IMF), whose feature energy, singular values and AR model parameters were computed as initial feature vectors. Secondly, ini tial feature vectors were mapped into a higherdimensional space with KPCA, and the kemel principal components were extracted as new feature vectors, which used as the input of LSSVM for fault classification. The experimental results show the KPCALSSVM method improves LSSVMs classification performance by KPCA obtaining additional discriminative information, and has better generalization than direct LSSVM method, and can identify rolling bearing fault patterns more accurately．
YANG Xian-Yong, ZHOU Xiao-Jun, ZHANG Wen-Bin, YANG Fu-Chun. Rolling bearing fault diagnosis based on local wave method
and KPCA-LSSVM. J4, 2010, 44(8): 1519-1524.
［1］ SUYKENS J A K,VANDEWALLE J.Least squares support vectors machine classifiers ［J］.Neural Processing Letters, 1999, 9(3): 293300．
［2］ 甘良志, 孙宗海, 孙优贤. 稀疏最小二乘支持向量机［J］. 浙江大学学报：工学版, 2007, 41(2): 245248．
GAN Liangzhi, SUN Zonghai, SUN Youxian. Sparse least squares support vector machine ［J］. Journal of Zhejiang University：Engineering Science, 2007,41(2): 245248．
［3］ OJEDA F, SUYKENS J A K, MOOR B D. Low rank updated LSSVM classifiers for fast variable selection ［J］. Neural Networks, 2008, 21(2/3): 437449．
［4］ 康海英, 栾军英, 郑海起, 等. 基于阶次跟踪和经验模态分解的滚动轴承包络解调分析［J］. 机械工程学报, 2007, 43(8): 119122．
KANG Haiying, LUAN Junying, ZHENG Haiqi, et al. Envelope demodulation analysis of bearing based on order tracking and Empirical mode decomposition ［J］. Chinese Journal of Mechanical Engineering, 2007, 43(8): 119122．
［5］ 张海勇. 一种新的非平稳信号分析方法—局域波分析［J］. 电子与信息学报, 2003, 25(10): 13271333．
ZHANG Haiyong. A new method for analyzing nonstation ary signallocal wave analysis ［J］. Journal of Electronics and Information Technology, 2003, 25(10): 13271333．
［6］ XU Yong, ZHANG David, SONG Fengxi, et al. A method for speeding up feature extraction based on KPCA ［J］. Neurocomputing, 2007, 70(46): 10561061．
［7］ HUANG N E, SHEN Z, STEVEN R L, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis ［J］. Proceedings of the Royal Society of London, A. 1998, 454(19): 903995．
［8］ 孙晖, 朱善安. 基于自适应滤波的滚动轴承故障诊断研究［J］. 浙江大学学报：工学版, 2005, 39(11): 17461749．
SUN Hui, ZHU Shanan. Rolling bearing fault diagnosis based on adaptive filtering ［J］. Journal of Zhejiang University：Engineering Science, 2005, 39(11): 17461749．
［9］ RILLING G, FLANDRIN P, GONCALVES P. On empirical mode decomposition and its algorithms ［C］∥IEEE EURASIP Workshop on NSIP03. Grado, Italy:IEEE, 2003: 811．
［10］ SCHLGL A. A comparison of multivariate autoregressive estimators ［J］. Signal Processing, 2006, 86: 24262429．
［11］ CHIH W H, CHIH J L.A comparison of methods for multiclass support vector machines ［J］.Neural Networks, 2002, 13(2): 415425．
［12］ Case Western Reserve University. Bearing data center ［EB/OL］. ［20080829］. http:∥www.eecs.cwru.edu/laboratory/bearing．
