Robust clustering of acoustic emission signals using the Kohonen network

Acoustic emission-based techniques are promising for nondestructive inspection of mechanical systems. For reliable automatic fault monitoring, it is important to identify the transient crack-related signals in the presence of strong time-varying noise and other interference. In this paper we propose the application of the Kohonen network for this purpose. The principal components of the short-time Fourier transforms of the data were applied input of the network. The clustering results confirm the capability of the Kohonen network for reliable source identification of acoustic emission signals, assuming enough care has been taken in implementing the training algorithm of the network.

[1]  D. Signorini,et al.  Neural networks , 1995, The Lancet.

[2]  D. Dornfeld.  Book reviewAcoustic emission, vol. 2: edited by J. R. Matthews; published by Gordon and Breach, New York, 1983; 168 pp.; price, U.S. $39.50 , 1985 .

[3]  Kevin M. Buckley,et al.  Signal processing for fault monitoring using acoustic emissions , 1999 .

[4]  L. Atlas,et al.  Feature extraction networks for dull tool monitoring , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[5]  Mostafa Kaveh,et al.  Detection and characterization of cracks for failure monitoring and diagnostics , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[6]  L. Cohen,et al.  Time-frequency distributions-a review , 1989, Proc. IEEE.

[7]  T. F. Drouillard.  A history of acoustic emission , 1996 .

[8]  Ahmed H. Tewfik,et al.  Acoustic emission transient detection based on linear model residuals , 1998, Ninth IEEE Signal Processing Workshop on Statistical Signal and Array Processing (Cat. No.98TH8381).

[9]  Jeffrey N. Schoess,et al.  CH-46 Sea Knight flight test results for the Rotor Acoustic Monitoring System (RAMS) , 1998, Smart Structures.

[10]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[11]  Ahmed H. Tewfik,et al.  Automatic fault monitoring using acoustic emissions , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  Ahmed H. Tewfik,et al.  Self-organizing map neural network for transient signal classification in mechanical diagnostics , 1999, NSIP.