Fault diagnosis of rolling element bearing using time-domain features and neural networks

Rolling element bearings are critical mechanical components in rotating machinery. Fault detection and diagnosis in the early stages of damage is necessary to prevent their malfunctioning and failure during operation. Vibration monitoring is the most widely used and cost-effective monitoring technique to detect, locate and distinguish faults in rolling element bearings. This paper presents an algorithm using feed forward neural network for automated diagnosis of localized faults in rolling element bearings. Normal negative log-likelihood value and kurtosis value extracted from time-domain vibration signals are used as input features for the neural network. Trained neural networks are able to classify different states of the bearing with 100% accuracy. The proposed procedure requires only a few input features, resulting in simple preprocessing and faster training. Effectiveness of the proposed method is illustrated using the bearing vibration data obtained experimentally.

[1]  M. Møller.  A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning , 1990 .

[2]  X. R. Li,et al.  Model-set adaptation in variable-structure MM estimation by hypothesis testing , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[3]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[4]  Asoke K. Nandi,et al.  Real-time classification of rotating shaft loading conditions using artificial neural networks , 1997, IEEE Trans. Neural Networks.

[5]  C. J. Li,et al.  Bearing Condition Monitoring by Pattern Recognition Based on Bicoherence Analysis of Vibrations , 1996 .

[6]  B. Datner,et al.  Analysis of Roller/Ball Bearing Vibrations , 1979 .

[7]  Monica Carfagni,et al.  Using force signals to monitor mechanical systems , 1989 .

[8]  B. Samanta,et al.  ARTIFICIAL NEURAL NETWORK BASED FAULT DIAGNOSTICS OF ROLLING ELEMENT BEARINGS USING TIME-DOMAIN FEATURES , 2003 .

[9]  Ibrahim Esat,et al.  ARTIFICIAL NEURAL NETWORK BASED FAULT DIAGNOSTICS OF ROTATING MACHINERY USING WAVELET TRANSFORMS AS A PREPROCESSOR , 1997 .

[10]  Tao Han,et al.  ART–KOHONEN neural network for fault diagnosis of rotating machinery , 2004 .

[11]  A. F. Stronach,et al.  Third-order spectral techniques for the diagnosis of motor bearing condition using artificial neural networks , 2002 .

[12]  Qiao Hu,et al.  Fault diagnosis of rotating machinery based on a new hybrid clustering algorithm , 2008 .

[13]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[14]  A C McCormick,et al.  Classification of the rotating machine condition using artificial neural networks , 1997 .

[15]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

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

[17]  D. W. Thomas,et al.  Bearing Fault Detection Using Adaptive Noise Cancelling , 1982 .

[18]  Bo Li,et al.  Detection of common motor bearing faults using frequency-domain vibration signals and a neural network based approach , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[19]  P. D. McFadden,et al.  Vibration monitoring of rolling element bearings by the high-frequency resonance technique — a review , 1984 .

[20]  Kenneth W. Bauer,et al.  Feature selection assessment and comparison using two saliency measures in an Elman recurrent neural network , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..