Artificial Neural Networks in Engineering - 1995

ISBN # 0-7918-0048-2

This paper is copyrighted by ASME Press. For reprints, please contact the ANNIE organizers at or ASME Press at The Americal Society of Mechanical Engineers, 345 East 47th Street, New York, NY 10017, USA.

Detecting Crack Growth in Metal Structures Using Temporal Processing and the Parametric Avalanche Stochastic Filter Neural Network

Harvey L. Bodine, Robert L. Dawes, Steven S. Henley
Martingale Research Corporation
Eric V. K. Hill
Aerospace Engineering Department, Embry-Riddle Aeronautical University, Daytona Beach, FL 32114-3900
In this paper we discuss the lessons learned from applying a novel neural network architecture, the Parametric Avalanche Stochastic Filter (PASF) to a complex nondestructive testing and signal processing problem. We addressed the problem of identifying crack growth in complex metal structures using acoustic emissions. To test our approach, multiple aluminum pressure vessels, simulating an aircraft pressure hull, were constructed, instrumented, and fatigued-cycled to destruction. By combining the PASF neural network and processing techniques from speech recognition, we achieved classification rates of up to 93% for crack growth in the presence of rivet noise, rubbing, membrane noise, and other clutter.


  1. Dawes, R., "Quantum Neurodynamics and the Parametric Avalanche", Final Technical Report MRC-NASA-89004, Martingale Research Corporation, August, (1989).
  2. Dawes, R., "The Parametric Avalanche Stochastic Filter", Final Technical Report MRC-NWSC-91A004, Martingale Research Corporation, April, (1993), Contract N60921-91-C0071.
  3. Carlyle, J.M., "Acoustic Emission Signal Analysis", Ph.D. Thesis, Brunei University, London (1982).
  4. Graham, L.J. and Elsley, R.K., "AE Source Identification by Frequency Spectral Analysis for an Aircraft Monitoring Application", Journal of Acoustic Emission, Vol. 2, Number 1/2, 47-55, (1983).
  5. McBride, S. L. and Maclachlan, J. W., "Acoustic Emission Monitoring of Aircraft Structures," Journal of Acoustic Emission, Vol. 4 (April 1985): 151-154.
  6. Scala, C.M. and Coyle, R.A., "Acoustic Emission Waveform Analysis to Identify Fatigue Crack Propagation in a Mirage Aircraft", Journal of Acoustic Emission, Vol. 6, Number 4, 249-256, (1987).

NOTE: This material was based on work sponsored by the National Science Foundation. The opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Back to NDT Page
Back to Publications

Skip to navigation