Application of a KPCA-KICA-HSSVM hybrid strategy in bearing fault detection

In Modern industrial, bearings are common in rotating machines. The processing data of bearings are nonlinear, and they usually are complicated distribution, which contains both Gaussian and non-Gaussian distributions. If using a single data distribution detection method, it will result in detecting performance degradation. To solve the possible monitoring difficulties of complicated distribution and nonlinear characteristics in industrial systems, this paper proposes a KPCA-KICA-HSSVM hybrid strategy. The proposed method uses KPCA, KICA and HSSVM to establish detection model, these models are work collaboratively in monitoring the real time process data to detect the possible faults. The proposed approach was tested and validated via a set of experimental data collected from a bearing test rig. Experimental results demonstrate the effectiveness of this approach.

[1]  Hao Wu,et al.  An adaptive confidence limit for periodic non-steady conditions fault detection , 2016 .

[2]  Robert P. W. Duin,et al.  Support Vector Data Description , 2004, Machine Learning.

[3]  Steven X. Ding,et al.  A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part II: Fault Diagnosis With Knowledge-Based and Hybrid/Active Approaches , 2015, IEEE Transactions on Industrial Electronics.

[4]  Nan Yang,et al.  Fault Isolation of Nonlinear Processes Based on Fault Directions and Features , 2014, IEEE Transactions on Control Systems Technology.

[5]  Tommy W. S. Chow,et al.  Motor Bearing Fault Diagnosis Using Trace Ratio Linear Discriminant Analysis , 2014, IEEE Transactions on Industrial Electronics.

[6]  Brigitte Chebel-Morello,et al.  Accurate bearing remaining useful life prediction based on Weibull distribution and artificial neural network , 2015 .

[7]  Wei Qiao,et al.  Bearing Fault Diagnosis for Direct-Drive Wind Turbines via Current-Demodulated Signals , 2013, IEEE Transactions on Industrial Electronics.

[8]  Yide Wang,et al.  Fault diagnosis method based on FFT-RPCA-SVM for Cascaded-Multilevel Inverter. , 2016, ISA transactions.

[9]  Hai Qiu,et al.  Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics , 2006 .

[10]  Steven X. Ding,et al.  A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part I: Fault Diagnosis With Model-Based and Signal-Based Approaches , 2015, IEEE Transactions on Industrial Electronics.

[11]  Fei Liu,et al.  Fault Detection and Diagnosis of Multiple-Model Systems With Mismodeled Transition Probabilities , 2015, IEEE Transactions on Industrial Electronics.

[12]  Bin Zhang,et al.  A Probabilistic Fault Detection Approach: Application to Bearing Fault Detection , 2011, IEEE Transactions on Industrial Electronics.

[13]  Xin Yao,et al.  The Impact of Diversity on Online Ensemble Learning in the Presence of Concept Drift , 2010, IEEE Transactions on Knowledge and Data Engineering.

[14]  Zhiyong Gao,et al.  A dual-parameter optimization KPCA method for process fault diagnosis , 2015, 2015 Annual Reliability and Maintainability Symposium (RAMS).

[15]  Bernhard Schölkopf,et al.  Kernel Principal Component Analysis , 1997, ICANN.

[16]  Elhoussin Elbouchikhi,et al.  Cascaded H-Bridge Multilevel Inverter System Fault Diagnosis Using a PCA and Multiclass Relevance Vector Machine Approach , 2015, IEEE Transactions on Power Electronics.

[17]  Masahiro Ono,et al.  Visualisation of the T cell differentiation programme by Canonical Correspondence Analysis of transcriptomes , 2014, BMC Genomics.

[18]  Mo-Yuen Chow,et al.  Neural-network-based motor rolling bearing fault diagnosis , 2000, IEEE Trans. Ind. Electron..

[19]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[20]  N. Zerhouni,et al.  Fault prognostic of bearings by using support vector data description , 2012, 2012 IEEE Conference on Prognostics and Health Management.