Abstract
This paper describes an inverse analysis method using hierarchical neural networks and computational mechanics, and its application to the quantitative nondestructive evaluation with the ultrasonic method. The present method consists of three subprocesses. First, by parametrically changing the location and size of a defect hidden in solid, elastic wave propagation in the solid is calculated with the dynamic finite element method. Second, the back-propagation neural network is trained using the calculated relationships between the defect parameters and the dynamic responses of solid surface. Finally, the trained network is utilized to determine appropriate defect parameters from some measured dynamic responses of solid surface. The accuracy and efficiency of the present method are discussed in detail through the identification of size and location of a defect hidden in solid.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Baker, A. R.; Windsor, C. G. 1989: The classification of defects from ultrasonic data using neural networks: The Hopfield method. NDT International 22: 97–105
Berry, D.; Upda, L.; Upda, S. S. 1991: Classification of ultrasonic signal via neural networks. In: Thompson, D. O.: Chimenti, D. E. (eds.); Review of Progress in Quantitative Nondestructive Evaluation 10A: 659–666. New York: Plenum Press
Blake, R. J.; Bond, L. J. 1990: Rayleigh wave scattering from surface features: wedges and down-steps. Ultrasonics 28: 214–228
Bond, L. J.; Taylor, J. 1991: Interaction of Rayleigh waves with a rib attached to a plate. Ultrasonics 29: 451–458
Brown, L. M.; DeNale, R. 1991: Classification of ultrasonic defect signatures using an artificial neural network. In: Thompson, D. O.; Chimenti, D. E. (eds.); Review of Progress in Quantitative Nondestructive Evaluation 10A: 705–712. New York: Plenum Press
Cichocki, A.; Unbehauen, R. 1993: Neural networks for optimization and signal processing. John Wiley and Sons
Damarla, T. R.; Karpur, P.; Bhagat, P. K. 1992: A self-learning neural net for ultrasonic signal analysis. Ultrasonics 30: 317–324
Elman, J. L.; Zipser, D. 1988: Learning the hidden structure of speech. Journal of the Acoustical Society of America 83: 1615–1626
Funahashi, K. 1989: On the approximate realization of continuous mappings by neural networks. Neural Networks 2: 183–192
Galdos, A.; Okuda, H.; Yagawa, G. 1990: Finite element simulation of ultrasonic wave propagation in pipe and pressure vessel walls. Finite Elements in Analysis and Design 7: 1–13
Georgiou, G. A.; and Bond, L. J. 1987: Quantitave studies in ultrasonic wave-defect interaction. Ultrasonics 25: 328–334
Grabec, I.; Sachse, W. 1989. Application of an intelligent signal processing system to accoustic emission analysis. Journal of the Acoustical Society of America 85: 1226–1235
Harumi, K. 1986: Computer simulation of ultrasonics in a solid. NDT International 19: 315–332
Hirao, M.; Fukuoka, H.; Miura, Y. 1982: Scattering of Rayleigh surface waves by edge cracks: numerical simulation and experiment. Journal of the Acoustical Society of America 72: 602–606
Hirose, Y.; Yamashita, K.; Hijiya, S. 1991: Back-propagation algorithm which varies the number of hidden units. Neural Networks 4: 61–66
Homma, K.; Yuki, H. 1991: AE source analysis by means of a neural network (1st report): Calculations of artificial waveforms. Transactions of the Japan Society of Mechanical Engineers 57A: 1916–1921 (in Japanese)
Kitano, H. 1990: Designing neural networks using genetic algorithms with graph generation system. Complex Systems 4: 461–476
Kitahara, M.; Achenbach, J. D.; Guo, Q. C.; Peterson, M.; Ogi, T.; Notake, M. 1991: Depth determination of surface-breaking cracks by a neural network. In: Thompson, D. O.; Chimenti, D. E. (eds.), Review of Progress in Quantitative Nondestructive Evaluation 10A: 698–696. New York: Plenum Press
Kubo, S. 1992: Classification of inverse problems arising in field problems and their treatment. In: Proceedings of IUTAM Symposium on Inverse Problems in Engineering Mechanics, 51–60. Springer-Verlag
Lewis, T. G.; El-Rewini, H. 1992: Introduction to Parallel Computing. Prentice-Hall
Ludwig, R.; Lord, W. 1985: A finite element formulation for ultrasonic NDT modeling. In: Thompson, D. O.; Chimenti, D. E. (eds.); Review of Progress in Quantitative Nondestructive Evaluation 4A: 37–43. New York: Plenum Press
Ludwig, R.; Lord, W. 1988: A finite-element study of ultrasonic wave propagation and scattering in an Alminum block. Materials Evaluation 46: 108–113
Mann, J. M.; Schmerr, L. W.; Moulder, J. C. 1991: Neural network inversion of uniform-field eddy current data. Materials Evaluation 49: 34–39
Ogi, T.; Notake, M.; Yabe, Y.; Kitahara, M. 1991: Application of neural network to classification of defects: basic study of weights. In: Thompson, D. O.; Chimenti, D. E. (eds.) Review of Progress in Quantitative Nondestructive Evaluation 10A: 683–688. New York: Plenum Press
Peretto, P. 1992: An Introduction to the Modeling of Neural Networks. Cambridge U.P.
Pratt, D.; Sansalone, M. 1991: The use of a neural network for automating impact-echo signal interpretation. In: Thompson, D. O.; Chimenti, D. E. (eds.); Review of Progress in Quantitative Nondestructive Evaluation 10A: 667–674 New York: Plenum Press
Rumelhart, D. E.; McClelland, J. L.; the PDP research group 1986: Parallel distributed processing: Explorations in the microstructure of cognition. Vol. 1: Foundations. MIT press
Sietsma, J.; Dow, R. J. F. 1991: Creating artificial neural networks that generalize. Neural Networks 4: 67–79
Song, S. J.; Schmerr, Jr, L. W. 1991: Ultrasonic flaw classification in weldments using neural networks. In: Thompson, D. O.; Chimenti, D. E. (eds.), Review of Progress in Quantitative Nondestructive Evaluation 10A: 697–704 New York: Plenum Press
Thompson, R. B. 1983: Quantitative ultrasonic nondestructive evaluation methods. Trans. ASME, Journal of Applied Mechanics 50: 1191–1201
Thomsen, J. J.; Lund, K. 1991: Quality control of composite materials by neural network analysis of ultrasonic power spectra. Materials Evaluation 49: 594–600
Upda, L.; Upda, S. S. 1990: Eddy current defect characterization using neural networks. Materials Evaluation 48: 342–347
Wu, X.; Ghaboussi, J.; Garrett, Jr, J. H. 1992: Use of neural networks in detection of structural damage. Computers and Structures 42: 649–659
Yagawa, G.; Yoshimura, S.; Mochizuki, Y.; Ohishi, T. 1993: Identification of crack shape hidden in solid by means of neural network and computational mechanics. In: Proceedings of IUTAM Symposium on Inverse Problems in Engineering Mechanics, 213–222. Springer-Verlag
Yagawa, G.; Yoshioka, A.; Yoshimura, S.; Soneda, N. 1993: A parallel finite element method with supercomputer network. Computers and Structures 47: 407–418
Yoshimura, S.; Yagawa, G. 1993: Inverse analysis by means of neural network and computational mechanics (Its application to structural identification of vibrating plates). In: Inverse Problems (Specialized Monograph of International Conference on Computational Engineering Science (ICES'92)), 184–193. Atlanta Technology Publications
Author information
Authors and Affiliations
Additional information
Communicated by S. N. Atluri, 20 September 1994
This work is financially supported by the Grant-in-Aid for the scientific research of the Ministry of Education, Japan.
Rights and permissions
About this article
Cite this article
Oishi, A., Yamada, K., Yoshimura, S. et al. Quantitative nondestructive evaluation with ultrasonic method using neural networks and computational mechanics. Computational Mechanics 15, 521–533 (1995). https://doi.org/10.1007/BF00350265
Issue Date:
DOI: https://doi.org/10.1007/BF00350265