The NRSF-SVM based method for nonlinear rotor bearing fault diagnosis

The fault diagnosis scheme of the rotor bearing system often conducted by using either signal analysis approach or modeling method. In practice, the structure of the rotor bearing system is complex and contains many nonlinear factors. Therefore, it is hard to use the model-based method for fault detection. Thus, signal analysis approach is more efficient. In the signal analysis approach, frequency response function is widely applied. However, the existing analyzing methods of frequency response function have some limitations, such as multidimensional property. Thus, in this study, the concept of Nonlinear Response Spectrum Function (NRSF) is proposed to solve the problem. Finally, a simulation is conducted to identify the multi-fault rotor bearing systems by the proposed NRSFs feature and Support Vector Machine (SVM) classifier, showing that the NRSF-SVM approach has an excellent performance in fault identification of rotor bearing system.

[1]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[2]  S. Billings,et al.  Output Frequency Characteristics of Nonlinear Systems , 1996 .

[3]  S. Sathiya Keerthi,et al.  Convergence of a Generalized SMO Algorithm for SVM Classifier Design , 2002, Machine Learning.

[4]  A. W. Lees,et al.  Model based Identification of Rotating Machines , 2009 .

[5]  Chen Guo,et al.  Nonlinear dynamic response analysis of unbalance-rubbing coupling faults of rotor-ball bearing-stator coupling system , 2007 .

[6]  Xiaoyuan Zhang,et al.  Multi-fault diagnosis for rolling element bearings based on ensemble empirical mode decomposition and optimized support vector machines , 2013 .

[7]  V. Rai,et al.  Bearing fault diagnosis using FFT of intrinsic mode functions in Hilbert-Huang transform , 2007 .

[8]  Sun Yi,et al.  On the non-linear dynamic behavior of a rotor–bearing system , 2004 .

[9]  Hao Tang,et al.  Fault diagnosis approach based on Volterra models , 2010 .

[10]  S. A. Billings,et al.  Energy transfer properties of non-linear systems in the frequency domain , 2004 .

[11]  Xu Jian-feng.  The GFRF Identification Model of Radar Fuze Receiver , 2004 .

[12]  Kong Xiangyu,et al.  Study on GFRF modeling method for nonlinear system , 2004, Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788).

[13]  K. Worden,et al.  Associated Linear Equations for Volterra operators , 2005 .

[14]  Christophe Pierre,et al.  Full Three-Dimensional Rotor/Stator Interaction Simulations in Aircraft Engines With Time-Dependent Angular Speed , 2016 .

[15]  T. K. Kundra,et al.  Model based online diagnosis of unbalance and transverse fatigue crack in rotor systems , 2004 .

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

[17]  Nalinaksh S. Vyas,et al.  Nonlinear Parameter Estimation in Rotor-Bearing System Using Volterra Series and Method of Harmonic Probing , 2003 .

[18]  Anoushiravan Farshidianfar,et al.  Rolling element bearings multi-fault classification based on the wavelet denoising and support vector machine , 2007 .

[19]  G. R. Tomlinson,et al.  Analytical description of the effects of system nonlinearities on output frequency responses: A case study , 2006 .

[20]  Stephen A. Billings,et al.  Output frequency response function of nonlinear Volterra systems , 2006, Autom..

[21]  Xin Xia,et al.  A novel method for fault diagnosis of hydro generator based on NOFRFs , 2015 .

[22]  Zhike Peng,et al.  Volterra-series-based nonlinear system modeling and its engineering applications: A state-of-the-art review , 2017 .

[23]  Bo-Suk Yang,et al.  Support vector machine in machine condition monitoring and fault diagnosis , 2007 .