Heuristic modeling using recurrent neural networks: simulated and real-data experiments

• Autorzy: Przystałka, P.
• Języki publikacji: EN

Abstrakty

EN

The focus of this paper is on the problems of system identification, process modeling and time series forecasting which can be met during the use of locally recurrent neural networks in heuristic modeling technique. However, the main interest of this paper is to survey the properties of the dynamic neural processor which is developed by the author. Moreover, a comparative study of selected recurrent neural architectures in modeling tasks is given. The results of experiments showed that some processes tend to be chaotic and in some cases it is reasonable to use soft computing models for fault diagnosis and control.

Słowa kluczowe

PL

sieć neuronowa; prognozowanie

EN

chaotic dynamic systems; recurrent neural network; soft computing learning algorithms; nonlinear system identification; time series forecasting

• Czasopismo: Computer Assisted Mechanics and Engineering Sciences
• Rocznik: 2007
• Tom: Vol. 14, No. 4
• Strony: 715-727
• Opis fizyczny: Bibliogr. 34 poz., rys.,tab., wykr.

Twórcy

autor

Przystałka, P.

• Silesian University of Technology Politechnika Śląska], Department of Fundamentals of Machinery Design, ul. Konarskiego 22, 44-100 Gliwice

Bibliografia

• [1] M. Ayoubi. Nonlinear dynamic systems identification with dynamic neural networks for fault diagnosis in technical processes. In Humans, Information and Technology, 3: 2120-2125, 2-5 October 1994.
• [2] M. Blanke, M. Kinnaert, J. Lunze, M. Staroswiecki. Diagnosis and Fault-Tolerant Control. Springer-Verlag, Berlin-Heidelberg, 2003.
• [3] W. Duch, J. Korbicz, L. Rutkowski, R. Tadeusiewicz. Sieci neuronowe, vol. 6 of Biocybernetyka i inzynieria biomedyczna. Akademicka Oficyna Wydawnicza EXIT, Warszawa, 2000.
• [4] K. Gobbak, H. Raghavendran, M. Tapas. Internal feedback neuron networks for modeling of an industrial furnace. In Neural Networks, pp. 700-705. IEEE, 1997.
• [5] S. Graziani, N. Pitrone, M.G. Xibilia, N. Barbalace. Improving monitoring of NOx emissions in refineries. In Instrumentation and Measurement Technology Conference, Proceedings of the 21st IEEE, vol. 1, pp. 594-597, 18-20 May 2004.
• [6] Martin T. Hagan, B.M. Menhaj. Training feedforward networks with the Marquardt algorithm. In Neural Networks, 5: 989-993, November 1994.
• [7] J.S. Heo, K.Y. Lee, R. Garduno-Ramirez. Multiobjective control of power plants using particle swarm optimization techniques. Energy Conversion, 21: 552-561, June 2006.
• [8] A. Jankowska. Modele emisji toksyn ze zbiorczego komina elektrowni — walidacja sygnałów wejściowych analizatora wirtualnego. In Pomiary Automatyka Kontrolna, vol. 9BIS, pp. 173-176, Sept. 2005.
• [9] P.S. Jarvis, I.D. Wilson, S.E. Kemp. The application of a new attribute selection technique to the forecasting of housing value using dependence modelling. Neural Computing and Applications, 15(2): 136-153, April 2006.
• [10] A.J. Jones. New tools in non-linear modelling and prediction. Computational Management Science, 1(2): 109-149, July 2004.
• [11] A.J. Jones, D. Evans, S.E. Kemp. A note on the Gamma test analysis of noisy input/output data and noisy time series. Physica D: Nonlinear Phenomena, 229(1): 1-8, 2007.
• [12] S.E. Kemp, I.D. Wilson, J.A. Ware. A Tutorial on the Gamma Test. International Journal of Simulation Systems, Science and Technology, 6(1-2), January 2005.
• [13] J. Korbicz. Robust fault detection using analytical and soft computing methods. Bulletin of the Polish Academy of Sciences: Technical Sciences, 54(1): 75-88, 2006.
• [14] J. Korbicz, J.M. Kościelny, Z. Kowalczuk, W. Cholewa. Fault Diagnosis. Models, Artificial Intelligence, Applications. Springer-Verlag, Berlin-Heidelberg, 2004.
• [15] J.M. Kościelny, M. Bartys. Models in diagnostics of industrial processes. Diagnostyka, 30: 291-296, 2004.
• [16] P.G. Luebbers, A.S. Pandya. Neural networks for process control. In Systems, Man and Cybernetics, pp. 393-398, Chicago, IL, USA, October 1992.
• [17] MAT LAB and Simulink for Technical Computing. The MathWorks.http://www.mathworks.com/access/helpdesk/help/toolbox/nnet/, 2007.
• [18] W. Moczulski. Methodology of heuristic modelling of dynamie objects and processes for diagnostics and control. In Recent Developments in Artificial Intelligence Methods, pages 123-126, November 2005.
• [19] W. Moczulski. Metodyka heurystycznego modelowania obiektów i procesów dynamicznych w diagnostyce i sterowaniu. In Pomiary Automatyka Kontrolna, vol. 9BIS, pp. 39-42, Sept. 2005.
• [20] K.S. Narendra, K. Parthasarathy. Identification and control of dynamie systems using neural networks. In IEEE Transactions on Neural Networks, pp. 4-27, 1990.
• [21] E. Ott. Chaos w Układach Dynamicznych. Wydawnictwo Naukowo-Techniczne, Warszawa, 1997.
• [22] S.J. Ovaska, A. Kamiya, Y.Q. Chen. Fusion of soft computing and hard computing: computational structures and characteristic features. Systems, Man and Cybernetics, Part C: Applications and Reuiews, 36: 439-448, May 2006.
• [23] F. Pasemann. A simple chaotic neuron. Physica D, 104: 205-211, May 1997.
• [24] R.B.C. Prudencio, T.B. Ludermir. Neural network hybrid learning: Genetic algorithms and Levenberg-Marąuardt. CiteSeer. IST - Copyright Penn State and NEC, 2001.
• [25] P. Przystałka. Heuristic modeling of objects and processes using dynamie neural networks. Diagnostyka, 2/38: 15-18, 2006.
• [26] M.T. Rosenstein, J.J. Collins, C.J. De Luca. A practical method for calculating largest Lyapunov exponents from smali data sets. Physica D, 65: 117-134, 1993.
• [27] M.T. Rosenstein, J.J. Collins, C.J. De Luca. Reconstruction expansion as a geometry-based framework for choosing proper delay times. Physica D, 73: 82-98, 1994.
• [28] M. Schlang, E. Broese, B. Feldkeller, O. Gramckow, M. Jansen, T. Poppe, C. Schaeffner, G. Soergel. Neural networks for process control in steel manufacturing. In International Conference on Acoustics, Speech, and Signal Processing, vol. 1, pp. 155-158. IEEE, 1997.
• [29] A. Stefansson, N. Konćar, A.J. Jones. A notę on the Gamma test. Neural Computing and Applications, 5(3): 131-133, Sept. 1997.
• [30] R. Tadeusiewicz. Sieci neuronowe jako wciąż jeszcze nie w pełni wykorzystane narzędzie diagnostyki. In Pro-ceedings of II International Congress of Technical Diagnostics, vol. 1, pp. 81-94, Warsaw, 2000.
• [31] F. Takens. Detecting strange attractors in turbulence. In D.A. Rand, L.S. Young, eds., Lecture Notes in Mathematics 898, pp. 366-381. Springer, Berlin, 1981.
• [32] A.C. Tsoi, A.D. Back. Locally recurrent globally feed-forward networks: a critical review of architectures. In Neural Networks, 5: 229-239, Mar 1994.
• [33]L. Tsungnan, CL. Giles, G. Bill, S.Y. Kung. A delay damage model selection algorithm for NARX neural networks. IEEE, 1999.
• [34] L.A. Zadeh. Fuzzy logie, neural networks and soft computing. Communications of the ACM, 37(3), March 1994.