Rapid Multi-Sensor Feature Fusion Based on Non-Stationary Kernel JADE for the Small-Amplitude Hunting Monitoring of High-Speed Trains

Joint Approximate Diagonalization of Eigen-matrices (JADE) cannot deal with non-stationary data. Therefore, in this paper, a method called Non-stationary Kernel JADE (NKJADE) is proposed, which can extract non-stationary features and fuse multi-sensor features precisely and rapidly. In this method, the non-stationarity of the data is considered and the data from multi-sensor are used to fuse the features efficiently. The method is compared with EEMD-SVD-LTSA and EEMD-JADE using the bearing fault data of CWRU, and the validity of the method is verified. Considering that the vibration signals of high-speed trains are typically non-stationary, it is necessary to utilize a rapid feature fusion method to identify the evolutionary trends of hunting motions quickly before the phenomenon is fully manifested. In this paper, the proposed method is applied to identify the evolutionary trend of hunting motions quickly and accurately. Results verify that the accuracy of this method is much higher than that of the EEMD-JADE and EEMD-SVD-LTSA methods. This method can also be used to fuse multi-sensor features of non-stationary data rapidly.

[1]  Wen Zefeng.  EFFECTS OF LATERAL DEFORMATIONS OF WHEELSET/TRACK ON CREEP FORCES OF WHEEL/RAIL , 2002 .

[2]  Bing He,et al.  Residual life prediction for ball bearings based on joint approximate diagonalization of eigen matrices and extreme learning machine , 2017 .

[3]  Klaus Nordhausen,et al.  Blind Source Separation Based on Joint Diagonalization in R: The Packages JADE and BSSasymp , 2017 .

[4]  James F. Peters,et al.  Multi-manifold LLE learning in pattern recognition , 2015, Pattern Recognit..

[5]  Siliang Lu,et al.  HKF-SVR Optimized by Krill Herd Algorithm for Coaxial Bearings Performance Degradation Prediction , 2020, Sensors.

[6]  Yunguang Ye,et al.  Small Amplitude Hunting Instability of High-speed Train Diagnosis Method Based on Modified Ensemble Empirical Mode Decomposition, Shannon Entropy and Least Square Support Vector Machine , 2016 .

[7]  Huajiang Ouyang,et al.  Multi-damage identification based on joint approximate diagonalisation and robust distance measure , 2017 .

[8]  Sang H. Lee,et al.  A Mixed Algorithm of PCA and LDA for Fault Diagnosis of Induction Motor , 2007, ICIC.

[9]  Hou,et al.  Study on Evaluation Methods for Lateral Stability of High-Speed Trains , 2012 .

[10]  Yuanyuan Pan,et al.  Machine Fault Classification Based on Local Discriminant Bases and Locality Preserving Projections , 2014 .

[11]  Oldrich Polach,et al.  Modelling of suspension components in a rail vehicle dynamics context , 2011 .

[12]  Robert X. Gao,et al.  Integration of EEMD and ICA for wind turbine gearbox diagnosis , 2014 .

[13]  Bin Li,et al.  Early Fault Detection of Machine Tools Based on Deep Learning and Dynamic Identification , 2019, IEEE Transactions on Industrial Electronics.

[14]  Yi Zhu,et al.  Analog Circuit Fault Diagnosis Using LDA and OAOSVM Approach , 2012 .

[15]  Guang Meng,et al.  A novel damage detection approach by using Volterra kernel functions based analysis , 2015, J. Frankl. Inst..

[16]  Huajiang Ouyang,et al.  A multi-sensor fusion framework for detecting small amplitude hunting of high-speed trains , 2018, Journal of Vibration and Control.

[17]  Qingbo He,et al.  A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification , 2015 .

[18]  A. D. Pater.  The approximate determination of the hunting movement of a railway vehicle by aid of the method of Krylov and Bogoljubov , 1961 .

[19]  Yude Bu,et al.  Stellar spectral subclasses classification based on Isomap and SVM , 2014 .

[20]  Qingbo He,et al.  Time–frequency manifold for nonlinear feature extraction in machinery fault diagnosis , 2013 .

[21]  Tao Wu,et al.  Fault Diagnosis of Bearings Based on KJADE and VNWOA-LSSVM Algorithm , 2019 .

[22]  Bing He,et al.  Feature fusion using kernel joint approximate diagonalization of eigen-matrices for rolling bearing fault identification , 2016 .

[23]  E. P. de Moura,et al.  Evaluation of principal component analysis and neural network performance for bearing fault diagnosis from vibration signal processed by RS and DF analyses , 2011 .

[24]  Damien Garcia,et al.  Color and Vector Flow Imaging in Parallel Ultrasound With Sub-Nyquist Sampling , 2018, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control.

[25]  Huajiang Ouyang,et al.  Feature recognition of small amplitude hunting signals based on the MPE-LTSA in high-speed trains , 2019, Measurement.

[26]  Jean-Franois Cardoso.  High-Order Contrasts for Independent Component Analysis , 1999, Neural Computation.

[27]  J. Benedetto,et al.  Nonlinear Dimensionality Reduction via the ENH-LTSA Method for Hyperspectral Image Classification , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[28]  Heiko Hoffmann,et al.  Kernel PCA for novelty detection , 2007, Pattern Recognit..

[29]  Jianhui Lin,et al.  Time-frequency processing of track irregularities in high-speed train , 2016 .

[30]  Qingbo He.  Vibration signal classification by wavelet packet energy flow manifold learning , 2013 .