Early Fault Diagnosis Using Laplacian Twin Least Squares Support Vector Machine

PDF(943 KB)

Journal of Vibration and Shock ›› 2017, Vol. 36 ›› Issue (16) : 85-92.

LI Feng1，TANG Baoping 2，GUO Yin3

Author information +

History +

Abstract

A novel early fault diagnosis method based on semi-supervised pattern recognition with Laplacian twin least squares support vector machine (Lap-TLSSVM) is proposed in this paper. In this method, the time-frequency domain feature set is first used to widely collect the feature information of various early faults. Then, the enhanced semi-supervised local Fisher discriminant Analysis (ESSLFDA) is utilized to reduce the high-dimensional time-frequency domain feature sets of training and testing samples to the low-dimensional eigenvectors with better category segregation. Finally, the low-dimensional eigenvectors of all samples are input into the introduced Lap-TLSSVM to conduct early fault diagnosis. In Lap-LSTSVM, the manifold regularization with large amounts of unlabeled data information is introduced to achieve semi-supervised learning. In addition, the twin objective functions of Lap-LSTSVM have only equality constraints and an efficient conjugate gradient (CG) algorithm is embedded in Lap-LSTSVM to solve the linear equations of objective functions for speeding up the training procedure. The proposed early fault diagnosis method has high diagnosis accuracy and computation efficiency even if the training sample set is small. Experimental results of early fault diagnosis on deep groove ball bearings confirm the effectiveness of the proposed method.

Key words

Rotating machinery / Manifold learning / Laplacian twin least squares support vector machine / Semi-supervised learning / Fault diagnosis

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

LI Feng1，TANG Baoping 2，GUO Yin3. Early Fault Diagnosis Using Laplacian Twin Least Squares Support Vector Machine[J]. Journal of Vibration and Shock, 2017, 36(16): 85-92

References

[1] TANG H F, CHEN J, DONG G M. Sparse representation based latent components analysis for machinery weak fault detection [J]. Mechanical Systems and Signal Processing, 2014, 46 (2): 373-388.
[2] WANG Y, XU G H, LIANG L, et al. Detection of weak transient signals based on wavelet packet transform and manifold learning for rolling element bearing fault diagnosis [J]. Mechanical Systems and Signal Processing, 2015, 54-55: 259-276.
[3] 王国彪，何正嘉，陈雪峰，等. 机械故障诊断基础研究“何去何从”[J]. 机械工程学报, 2013, 49(1)：63-72.
WANG Guo-biao, HE Zhen-gjia, CHEN Xue-feng, et al. Review of life prediction for mechanical major equipments [J]. Chinese journal of mechanical engineering, 2013, 49(1)：63-72.
[4] LI W, RUAN Q Q, WAN J. Dimensionality reduction using graph-embedded probability-based semi-supervised discriminant analysis [J]. Neurocomputing, 2014, 138:283-296.
[5] HUANG H, LI J W, LIU J M. Enhanced semi-supervised local Fisher discriminant analysis for face recognition [J]. Future Generation Computer Systems, 2012, 28 (1): 244-253.
[6] LI F, WANG J X, CHYU M K, et al. Weak fault diagnosis of rotating machinery based on feature reduction with Supervised Orthogonal Local Fisher Discriminant Analysis [J]. Neurocomputing, 2015, 168:505-519.
[7] LI F, CHYU M K, WANG J X, et al. Life grade recognition of rotating machinery based on Supervised Orthogonal Linear Local Tangent Space Alignment and Optimal Supervised Fuzzy C-Means Clustering [J]. Measurement, 2015, 73: 384-400.
[8] JIANG L, XUAN J P, SHI T L. Feature extraction based on semi-supervised kernel Marginal Fisher analysis and its application in bearing fault diagnosis [J]. Mechanical Systems and Signal Processing, 2013, 41: 113-126.
[9] ZHANG S W, LEI Y K, WU Y H. Semi-supervised locally discriminant projection for classification and recognition [J]. Knowledge-Based Systems, 2011, 24: 341-346.
[10] 张小丽，陈雪峰，李兵，等. 机械重大装备寿命预测综述[J]. 机械工程学报, 2011, 47 (11):100-116.
ZHANG Xiao-li, CHEN Xue-feng, LI Bing, et al. Review of life prediction for mechanical major equipments [J]. Journal of Mechanical Engineering, 2011, 47 (11):100-116.
[11] ANTONUCCI A, ROSA R D, GIUSTI A, et al. Robust classification of multivariate time series by imprecise hidden Markov models [J]. International Journal of Approximate Reasoning, 2015, 56: 249-263.
[12] ZOUHAL L M, DENOEUX T. An Evidence-Theoretic k-NN Rule with Parameter Optimization [J]. IEEE Transactions on Systems, Man, and Cybernetics, 1998, 28 (2): 263-271.
[13] 李锋，汤宝平，宋涛. 基于Shannon小波支持矢量机二级决策的故障诊断[J]. 机械工程学报, 2010, 46 (17): 42-47.
LI Feng, TANG Bao-ping, SONG Tao. Fault Diagnosis Based on the Second Class Decision Making through Shannon Wavelet Support Vector Machine [J]. Journal of Mechanical Engineering, 2010, 46 (17): 42-47.
[14] CHEN W J, SHAO Y H, DENG N Y, et al. Laplacian least squares twin support vector machine for semi-supervised classification [J]. Neurocomputing, 2014, 145: 465–476.
[15] BELKIN M, NIYOGI P, SINDHWANI V. Manifold regularization: a geometric framework for learning from labeled and unlabeled examples [J]. J. Mach. Learn. Res., 2006, 7: 2399-2434.