Multi-layered Bayesian Neural Networks for Simulation and Prediction Stress-Strain Time Series

• Autorzy: Krok A.
• Języki publikacji: EN

Abstrakty

EN

The aim of the paper is to investigate the differences as far as the numerical accuracy is concerned between feedforward layered Artificial Neural Networks (ANN) learned by means of Kalman filtering (KF) and ANN learned by means of the evidence procedure for Bayesian technique. The stress-strain experimental time series for concrete hysteresis loops obtained by the experiment of cyclic loading is presented as considered example.

Słowa kluczowe

EN

Bayesian neural network; Kalman filtering

• Wydawca: Instytut Łączności - Państwowy Instytut Badawczy
• Czasopismo: Journal of Telecommunications and Information Technology
• Rocznik: 2015
• Tom: nr 3
• Strony: 45--51
• Opis fizyczny: Bibliogr. 40 poz., rys.

Twórcy

autor

Krok A.

• Faculty of Physics, Mathematics and Computer Science, Tadeusz Kościuszko Cracow University of Technology, Warszawska st 24, 31-155 Cracow, Poland

Bibliografia

• [1] S. Haykin, Neural Networks: A Comprehensive Foundation, 2nd ed. Prentice Hall, 1999.
• [2] T. Bayes, “An Essay towards solving a problem in the doctrine of chances”, Philosoph. Trans. of the Royal Society of London, vol. 53, pp. 370–418, 1763.
• [3] D. J. C. MacKay, Information Theory, Inference and Learning Algorithms. Cambridge University Press, 2003.
• [4] T. Auld, A. W. Moore, and S. F. Gull, “Bayesian neural networks for Internet traffic classification”, IEEE Trans. Neural Netw., vol. 18, no. 1, pp. 223–239, 2007.
• [5] D. MacKay, “Bayesian neural networks and density networks”, Nuclear Instrum. and Methods in Phys. Res. Section A, vol. 354, no. 1, pp. 73–80, 1995.
• [6] J. Lampinen and A. Vehtari,“ Bayesian approach for neural networks and case studies”, Neural Netw., vol. 14, no. 3, pp. 257–274, 2001.
• [7] G. Büyüközkan, G. Kayakutlu, and I. S. Karakadılar, “Assessment of lean manufacturing effect on business performance using Bayesian Belief Networks”, Expert Syst. with Applications, vol. 42, no. 19, pp. 6539–6551, 2015.
• [8] I. Kononenko, “Bayesian neural networks”, Biolog. Cybernetics, vol. 61, pp. 361–370, 1989.
• [9] O. Kocadaǧli and B. Aşıkgil, “Nonlinear time series forecasting with Bayesian neural networks”, Expert Syst. with Applications, vol. 41, no. 15, pp. 6596–6610, 2014.
• [10] A. A. Zaidan et al., “Image skin segmentation based on multi-agent learning Bayesian and neural network”, Engin. Appl. of Artif. Intell., vol. 32, pp. 136–150, 2014.
• [11] M. N. A. Khan, “Performance analysis of Bayesian networks and neural networks in classification of file system activities”, Comp.& Secur., vol. 31, no. 4, pp. 391–401, 2012.
• [12] H. S. Hippert and J. W. Taylor, “An evaluation of Bayesian techniques for controlling model complexity and selecting inputs in a neural network for short-term load forecasting”, Neural Netw., vol. 23, no. 3, pp. 386–395, 2010.
• [13] Ch. K. I. Williams and F. Vivarellia, “Comparing Bayesian neural network algorithms for classifying segmented outdoor images”, Neural Netw., vol. 14, no. 4–5, pp. 427–437, 2001.
• [14] M. Hamimid, S. M. Mimoune, and M. Feliachi, “Minor hysteresis loops model based on exponential parameters scaling of the modified Jiles–Atherton model”, Physica B: Condensed Matter, vol. 407, no. 13, pp. 2438–2441, 2012.
• [15] A. Ganczarski and L. Barwacz, “Concept of continuous damage deactivation in modelling of low cycle fatigue”, in Proc. Int. Conf. on Fracture ICF11 2005, Turin, Italy, 2005.
• [16] K. Chwastek, “Modelling hysteresis loops in thick steel sheet with the dynamic Takács model”, Physica B: Condensed Matter, vol. 407, no. 17, pp. 3632–3634, 2012.
• [17] M. Al Janaideh, “A time-dependent stop operator for modeling a class of singular hysteresis loops in a piezoceramic actuator”, Physica B: Condensed Matter, vol. 413, pp. 100–104, 2013.
• [18] A. P. S. Baghel, A. Gupta, K. Chwastek, and S. V. Kulkarni, “Comprehensive modelling of dynamic hysteresis loops in the rolling and transverse directions for transformer laminations”, Physica B: Condensed Matter, vol. 462, pp. 86–92, 2015.
• [19] I. Kucuk, “Prediction of hysteresis loop in magnetic cores using neural network and genetic algorithm”, J. Magnetism and Magnetic Materials, vol. 305, no. 2, pp. 423–427, 2006.
• [20] R. Dong, Y. Tan, H. Chen, and Y. Xie, “A neural networks based model for rate-dependent hysteresis for piezoceramic actuators”, Sensors and Actuators A: Physical, vol. 143, no. 2, pp. 370–376, 2008.
• [21] A. Nouicer, E. Nouicer, and F. Mouloudc, “A neural network for incorporating the thermal effect on the magnetic hysteresis of the 3F3 material using the Jiles–Atherton model”, J. Magnetism and Magnetic Materials, vol. 373, pp. 240–243, 2015.
• [22] V. Wolfs and P. Willems, “Development of discharge-stage curves affected by hysteresis using time varying models, model trees and neural networks”, Environ. Model. & Softw., vol. 55, pp. 107–119, 2014.
• [23] X. Zhang, Y. Tan, and M. Su, “Modeling of hysteresis in piezoelectric actuators using neural networks”, Mechan. Syst. and Sig. Process., vol. 23, no. 8, pp. 2699–2711, 2009.
• [24] R. E. Kalman, “A new approach to linear filtering and prediction problems”, Trans. ASME – J. of Basic Engin., vol. 82, series D, pp. 35-45, 1960.
• [25] L. Prechelt, “Adaptive parameter pruning in neural networks”, ICSI Tech. Rep. TR-95-009, Berkeley, CA, USA, 1995.
• [26] S. Haykin, Ed., Kalman Filtering and Neural Networks. New York: Wiley, 2001.
• [27] C. M. Bishop, Neural Networks for Pattern Recognition. Oxford University Press, 1995.
• [28] R. M. Neal, “Bayesian Learning for Neural Networks”, Lecture Notes in Statistics, vol. 118. Springer, 1996.
• [29] A. Krok, “Analysis of mechanics of structures and material problems applying artificial neural networks learnt by means of Kalman fltering”, Ph.D. thesis, Institute of Computer Mhetods in Civil Engineering, Cracov University of Techology, 2007 (in Polish).
• [30] A. Krok, “The development of Kalman filter learning technique for artificial neural networks”, J. Telecom. Inform. Technol., no. 4, pp. 16–21, 2013.
• [31] A. Krok and Z.Waszczyszyn, “Kalman filtering for neural prediction of response spectra from mining tremors”, in Proc. Int. Conf. Artif. Intell. AI-METH 2004, Gliwice, Poland, 2004.
• [32] A. Krok and Z. Waszczyszyn, “Simulation of building loops for a superconductor using neural networks with Kalman filtering”, Comp. Assisted Mechanics and Engin. Sci., vol. 13, pp. 575–582, 2006.
• [33] A. Krok and Z.Waszczyszyn, “Kalman filtering for neural prediction of response spectra from mining tremors”, Computers and Structures, vol. 85, no. 15–16, pp. 1257–1263, 2007.
• [34] A. Krok, “Simulation of concrete hysteresis loops using gaussian processes calibrated by genetic algorithm”, in Proc. Int. Symp. IPM on Inverse Problems in Mechanics of Struc. Mater., Rzeszow-Łańcut, Poland, 2009.
• [35] H. H. Thodberg, “Ace of bayes: application of neural networks with pruning”, Tech. Rep. 1132E, The Danish Meat Research Institute, 1993.
• [36] A. Ganczarski and L. Barwacz, “Low cycle fatigue based on unilateral damage ewolution”, Int. J. Damage Mech., vol. 16, no. 2, pp. 159–177, 2007.
• [37] A. Ylinen, “A method of determining the buckling stress and the required cross-sectional area for centrally loaded straight columns in the elastic and inelastic range”, Int. Assoc. for Bridges and Struct. Engin., vol. 16, pp. 529–550, 1956.
• [38] I. T. Nabney, Netlab: Algorithms for Pattren Recognition. Springer, 2002.
• [39] H. H. Thodberg, “Review of Bayesian neural networks with application to near infrared spectroscopy”, IEEE Trans. Neural Netw., vol. 7, no. 1, pp. 56–72, 1996.
• [40] Y. S. Othmana et al., “Frequency based hysteresis compensation for piezoelectric tube scanner using Artificial Neural Networks”, Procedia Engin., vol. 41, pp. 757–763, 2012.