Farah Hani NORDIN, Farrukh Hafiz NAGI, Aidil Azwin Zainul ABIDIN

Comparison study of computational parameter values between LRN and NARX in identifying nonlinear systems

To determine the nonlinear autoregressive model with exogenous inputs (NARX) parameter values is not an easy task, even though NARX is reported to successfully identify nonlinear systems. Apart from the activation functions, number of layers, layer size, learning rate, and number of epochs, the number of delays at the input and at the feedback loop need to also be determined. The layer recurrent network (LRN) is seen to have the potential to outperform NARX. However, not many papers have reported on using the LRN to identify nonlinear systems. Therefore, it is the aim of this paper to investigate and analyze the parametric evaluation of the LRN and NARX in identifying 3 different types of nonlinear systems. From the 3 nonlinear systems, the satellite's attitude state space is more complex compared to the sigmoid and polynomial equations. To ensure an unbiased comparison, a general guideline is used to select the parameter values in an organized manner. The LRN and NARX performance is analyzed based on the training and architecture parameters, mean squared errors, and correlation coefficient values. The results show that the LRN outperformed NARX in training quality, needs equal or fewer parameters that need to be determined through heuristic processes and equal or lower number of epochs, and produced a smaller training error compared to NARX, especially when identifying the satellite's attitude. This indicates that the LRN has the capability of identifying a more complex and nonlinear system compared to NARX.

Anahtar Kelimeler:

Layer recurrent network, nonlinear autoregressive with exogenous inputs, nonlinear system identification, recurrent neural network

Comparison study of computational parameter values between LRN and NARX in identifying nonlinear systems

Keywords:

Layer recurrent network, nonlinear autoregressive with exogenous inputs, nonlinear system identification, recurrent neural network,

PDF

___

H. Lee, Y. Park, K. Mehrotra, C. Mohan, S. Ranka, “Nonlinear system identiﬁcation using recurrent networks”, IEEE International Joint Conference on Neural Networks, Vol. 3, pp. 2410–2415, 1991.
D. Psichogios, L. Ungar, “Nonlinear internal model control and model predictive control using neural networks”, Proceedings of the 5th IEEE International Symposium on Intelligent Control, pp. 1082–1087, 1990.
R. Gao, A. O’Dwyer, E. Coyle, “Model predictive control of CSTR based on local model networks”, Proceedings of the Irish Signals and Systems Conference, pp. 397–402, 2002.
D. Soloway, P.J. Haley, “Neural generalized predictive control: A Newton-Raphson implementation”, Proceedings of the 11th IEEE International Symposium on Intelligent Control, pp. 277–282, 1996.
M. Basso, L. Giarre, S. Groppi, G. Zappa, “NARX models of an industrial power plant gas turbine”, IEEE Transactions on Control Systems Technology, Vol. 13, pp. 599–604, 2005.
H. Luo, S. Puthusserypady, “NARX neural networks for dynamical modelling of fMRI data”, Proceedings of the International Joint Conference on Neural Networks, pp. 542–546, 2006.
I. Maio, I. Stievano, F.G. Canavero, “NARX approach to black-box modelling of circuit elements”, Proceedings of the IEEE International Symposium on Circuits and Systems, Vol. 3, pp. 411–414, 1998.
H.P.H. Anh, N.H. Phuc, “Inverse neural MIMO NARX model identiﬁcation of nonlinear system optimized with PSO”, Proceedings of the 5th IEEE International Symposium on Electronic Design, Test and Applications, pp. 144–149, 2010.
H.P.H. Anh, “Inverse dynamic model identiﬁcation of 2-axes PAM robot arm using neural MIMO NARX model”, IEEE/ASME International Conference on Advanced Intelligent Mechatronics, pp. 1282–1287, 2009.
E. Pisoni, M. Farina, C. Carnevale, L. Piroddi, “Forecasting peak air pollution levels using NARX models”, Engineering Applications of Artiﬁcial Intelligence, Vol. 22, pp. 593–602, 2009.
R. Liutkeviˇcius, “Fuzzy Hammerstein model of nonlinear plant”, Nonlinear Analysis: Modelling and Control, Vol. 13, pp. 201–212, 2008.
K.K. Ahn, H.P.H. Anh, “Inverse double NARX fuzzy modeling for system identiﬁcation”, IEEE/ASME Transactions on Mechatronics, Vol. 15, pp. 136–148, 2010.
T.N. Lin, C. Giles, B.G. Horne, S.Y. Kung, “A delay damage model selection algorithm for NARX neural networks”, IEEE Transactions on Signal Processing, Vol. 45, pp. 2719–2730, 1997.
J.L. Elman, “Finding structure in time”, Cognitive Science, Vol. 14, pp. 179–211, 1990.
H. Demuth, M. Beale, M. Hagan, Neural Network’s Toolbox 5 User’s Guide, MathWorks, 2007.
M. Saggar, T. Mericli, S. Andoni, R. Miikkulainen, “System identiﬁcation for the Hodgkin-Huxley model using artiﬁcial neural networks”, International Joint Conference on Neural Networks, pp. 2239–2244, 2007.
F.H. Nordin, F.H. Nagi, “Layer-recurrent network in identifying a nonlinear system”, International Conference on Control, Automation and Systems, pp. 387–391, 2008.
M.J. Sidi, Spacecraft Dynamics & Control: A Practical Engineering Approach, Cambridge, Cambridge University Press, 1997.
F.C. Moon, Applied Dynamics: With Applications to Multibody and Mechatronic Systems, New York, Wiley, 1998. S. Thongchet, S. Kuntanapreeda, “A fuzzy-neural bang-bang controller for satellite attitude control”, Journal of King Mongkut’s Institute of Technology North Bangkok, Vol. 11, pp. 11–17, 2001.
W. Gang, L. Hua, K.F. Man, S. Kwong, “A robust nonlinear system identiﬁcation algorithm based on homotopy theory”, IEEE International Conference on Systems, Man, and Cybernetics, Computational Cybernetics and Simulation, Vol. 2, pp. 1313–1316, 1997.
X. Li, T. Chen, “System identiﬁcation using a new dynamic neural network”, Proceedings of International Joint Conference on Neural Networks, Vol. 3, pp. 2363–2366, 1993.
R.O. Duda, P.E. Hart, D.G. Stork, Pattern Classiﬁcation, New York, Wiley, 2001.
J.D. Villiers, E. Barnard, “Backpropagation neural nets with one and two hidden layers”, IEEE Transactions on Neural Networks, Vol. 4, pp. 136–141, 1992.
S. Haykin, Neural Networks: A Comprehensive Foundation, New Jersey, Prentice Hall, 1999.
B.G. Kermani, M.W. White, H.T. Nagle, “A new method in obtaining a better generalization in artiﬁcial neural networks”, Proceedings of the 16th Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering Advances: New Opportunities for Biomedical Engineers, pp. 1119–1120, 1994.