Abstract
High level loads in the acoustic frequency range are the reason for the aviation constructions’ elements fatigue destruction and death of the airborne equipment. Acoustic loads have most influence on the thin-walled elements of the aircraft construction. In this work we define the task of fault detection and present the method for solving this problem. Also we present the experiment results of the vibroacoustic system state detection using neural network method for the small size sample set situation.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Hamamoto, Y., Uchimura, S., Kanaoka, T., Tomita, S.: Evaluation of artificial neural network classifiers in small sample size situations. In: Pros. Int. It. Conf. Neural Networks, Nagoya, October 25-29, pp. 1731–1735 (1993)
Archer Norman, P., Shonhong, W.: Learning bias in neural networks and an approach to controlling its effects in monotonic classifications. IEEE Trans. Pattern Anal. And Mach. Intell. 15(9), 962–966 (1993)
Rosenblatt, F.: The principles of neurodynamics. Mir (1965)
Galushkin, A.I.: The synthesis of multilayer systems for pattern recognition. M. Energy (1974)
Galushkin, A.I.: Neural Network Theory. M. IPRGR (2000)
Baranov, S.N., Kuravsky, L.S.: Acoustic vibrations in thin-walled constructions: Modeling, optimization and analysis. RusAvia (2001)
Galushkin, A.I.: Neural networks and the problem of small training set. In: V All-Russia conference Neurocomputers and their applications (1999)
Zador Anthony, M., Pearlmutter Barak, A.: VC dimension of an integrate-and-fireneuron model. Neural Comput. 8(3), 624 (1996)
Bartlett Peter, L., Williamson Robert, C.: The VC dimension and pseudo dimension of twolayer neural networks with discrete inputs. Neural Comput. 8(3), 625–628 (1996)
Hole, A.: Some variants on a theorem of Vapnik. Prepr. Ser. Pure Math., Inst. Math. Univ. Oslo 30, 1–11 (1995)
Arne, H.: Vapnik-Chervonenkis generalization bounds for real valued neural networks. Neural Comput. 8(6), 1277–1299 (1996)
Goutam, C., Norio, S., Shoichi, N.: Optimization of overtraining and overgeneralization. In: Proc. Int. Jt. Conf. Neural Networks, Nagoya, October 25-29, vol. 3, pp. 2257–2262 (1993)
Tzafestas, S.G., Dalianis, P.J., Anthopoulos, G.C.: On the overtraining phenomenon of backpropagation neural networks. Math. and Comput. Simul. 40(5-6), 507–520 (1996)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Galushkin, A.I., Katsin, A.S., Korobkova, S.V., Kuravsky, L.S. (2004). Neural Network Classification Algorithm for the Small Size Training Set Situation in the Task of Thin-Walled Constructions Fatigue Destruction Control. In: Pal, N.R., Kasabov, N., Mudi, R.K., Pal, S., Parui, S.K. (eds) Neural Information Processing. ICONIP 2004. Lecture Notes in Computer Science, vol 3316. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30499-9_199
Download citation
DOI: https://doi.org/10.1007/978-3-540-30499-9_199
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23931-4
Online ISBN: 978-3-540-30499-9
eBook Packages: Springer Book Archive