Bayesian neural networks for prediction of response spectra

• Autorzy: Waszczyszyn, Z. , Słoński, M.
• Języki publikacji: EN

Abstrakty

EN

Standard artificial neural networks and Baycsian neural networks (BNNs) are briefly discussed on example of a simple feed-forward layered neural networks (FLNN). Main ideas of the Baycsian approach and basics of the applied BNNs are presented in short. A study case corresponds to prediction of Displacement Response Spectrum inside buildings at the basement level (DRSb). Data for network training and testing were adopted as DRS corresponding to the preprocessed accelerograms. They were taken from measurements in the Lcgnica-Gtogow Copperfield at monitored 5-storey buildings subjected to paraseismic excitations from explosives in nearby strip mines. Results of neural predictions by three NNs (standard FLNN trained by means of the conjugate gradient learning method, Simple Bayesian SBNN and Full Bayesian FBNN) are presented. The errors of predictions are on average on the level of 4% errors.

Słowa kluczowe

PL

sieć neuronowa; sprzężenie do przodu; funkcja gęstości prawdopodobieństwa; inżynieria lądowa; inżynieria budowlana

EN

feed-forward layered neural network; Bayesian neural network; probability density function; civil engineering; structural engineering; displacement response spectrum

• Czasopismo: Foundations of Civil and Environmental Engineering
• Rocznik: 2006
• Tom: No. 7
• Strony: 343-361
• Opis fizyczny: Bibliogr. 27 poz.

Twórcy

autor

Waszczyszyn, Z.

autor

Słoński, M.

• Cracow University of Technology, Faculty of Civil Engineering, Warszawska 24, 31-155 Kraków, Poland Tel.: +48 [phone] fax: +48 [phone],

Bibliografia

• 1. Basis for design of structures - Seismic actions on structures, ISO/DIS 3010, 2000.
• 2. Bishop Ch. M.: Neural Networks for Pattern Recognition, Clarendon Press, Oxford. 1995.
• 3. Ciesielski R.. Kuźniar K., Maciąg E., Tatara T.: Empirical formulae for fundamental natural periods of buildings with load bearing walls, Archives of Civil Engineering, 38, 4 (1992) 291-299.
• 4. Ciesielski R., Kuźniar K., Maciąg E., Tatara T: Damping of vibration in precast buildings with bearing concrete walls, Archives of Civil Eng- ineering, 41,3(1995)329-341.
• 5. Design of structures for earthąuake resistance. Eurocode 8, 2003.
• 6. Haykin S.: Neural Networks - A Comprehensive Foundation, Prentice-Hall, Upper Saddle River, NJ, USA, 1999.
• 7. Korbicz J., Obuchowicz A., Uciński D.: Artificial Neural Network -Foundations and Applications (in Polish), Akad. Ofic. Wyd PLJ, Warsaw, 1994.
• 8. Krok A., Waszczyszyn Z.: Neurai prediction of response spectra from mining tremors using recurrent laycred networks and Kalman filtering, in: Computational Fluid and Solid Mechanics, editor K.J. Bathe., Proc. 3rd MIT Conf., Elsevier, Amsterdam, 2005, 302-305.
• 9. Krok A,, Waszczyszyn Z.: Application of neural networks and Kalman filtering to simulation of displacement response spectra on buildings subjected to paraseismic excitations. in CDROM Proc. of Int. Symp. on Neural Networks and Soft Computing in Structural Eng., editors Z. Waszczyszyn . M. Słoński, ECCOMAS Thematic Conf., Cracow, 2005, (in preparation for publiting), 2007.
• 10. Kuźniar K.: BP Neural network computation of response spectra using a subpicture idea, in: Neural Networks and Soft Computing, editors L. Rutkowski, J.Kasprzyk, Physica-Verlag, A Springer-Yerlag Company, Heidelberg - New York, 2003, 760-765.
• 11. Kuźniar K.: Analysis of Yibrations of Medium-Height Buildings with Load Bearing Walls Subjected to Mining Tremors Using Neural Networks (in Polish), Monograph 310, Series of Civil Eng., Cracow Univ. of Technology, Poland, 2004.
• 12. Kuźniar K., Maciąg E., Waszczyszyn Z: Computation of response spectra from mining tremors using neural networks, Journal Soil Dynamics and Earthąuake Engineering, 25, 4 (2005) 331 -339.
• 13. Kuźniar K., Waszczyszyn Z.: Neural networks for the simulation and identification analysis of buildings subjected to paraseismic excitations. Ch. l lin: Intclligent Computational Paradigms in Earthquake Engineering, editor N. Lagaros (in print). Idea Group, Inc., Hershey, PA, USA, 2006.
• 14. Lampinen J., Yehtari A.: Bayesian approach for neural networks - review and case studies, Neural Networks, 14, 3(2001) 7-24.
• 15. MacKay D.J.C.: Bayesian Interpolation, Neural Computation, 4, 2(1991) 415-447.
• 16. MacKay D.J.C., Information Theory, Inference and Eearning Algorithms, Cambridge University Press, 2003.
• 17. Nabney I.T.: NETLAB - Algorithms for Pattcrn Recognition, Springer, Printed in Great Britain, 2nd printing with corrections, 2003.
• 18. Neal R.M.: Bayesian Training of Backpropagation Networks by the Hybrid Monte Carlo Method, Technical Report CRG-TR-92-1, Department of Computer Sciences, University of Toronto, 1992.
• 19. Neal R.M.: Bayesian Eearning for Neural Networks, Eecture Notes in Statistics, No.11, Springer, New York, 1996.
• 20. Neural Networks in the Analysis and Design of Structures, editor Z. Waszczyszyn, CISM Courses and Lectures No. 404. Springer, Wien - New York, 1999.
• 21. Neural Networks, Biocybernetics and Medical Engineering Vol.6, cditors W. Duch et al. (in Polish), Akad. Ofic. Wydawn. Exit, Warsaw, 2000.
• 22. Rojas R.: Neural Networks - A Systematic Introduction, Springcr Yerlag. Berlin, Germany, 1996.
• 23. Słoński M.: Prediction of concrete fatigue durability using Bayesian neural networks, Computer Assisted Mechanics and Engineering Sciences, 12, 2/3 (2005)259265-159.
• 24. Tipping M. E.: Bayesian inferencc: an introduction to principles and practice in machinę learning, in: Bayesian Learning: An Introduction to Principles and Practice in Machinę Learning , editors O. Bousąuet et al., Advanccd Lecturcs on Machinę Learning, Springer, 2004, 41-62.
• 25. Waszczyszyn Z.: Applications of Artificial Neural Networks in the Analysis of Civil Engineering Problems, Subsidy of the Foundation for Polish Science, No. 13/2001, Finał Report (in Polish), Inst. of Computer Methods in Civil Engineering, Cracow University of Technology, 2004.
• 26. Waszczyszyn Z., Ziemiański L.: Neural networks in the identification analysis of structural mechanics problems, in: Parameter Identification of Materials and Structures, editors Z. Mróz and G. Stavroulakis , CISM Courses and Lectures No.469, Springer, Wien - New York, 2005, 265-340.
• 27. Waszczyszyn Z., Ziemiański L.: Neurocomputing in the analysis of selected inverse problems of mechanics of Structures and materials, Computer