Nonlinear system identification with a real-coded genetic algorithm (RCGA)

Cherif, I. ; Fnaiech, F.

International Journal of Applied Mathematics and Computer Science, Tome 25 (2015) no. 4, pp. 863-875.

Résumé

This paper is devoted to the blind identification problem of a special class of nonlinear systems, namely, Volterra models, using a real-coded genetic algorithm (RCGA). The model input is assumed to be a stationary Gaussian sequence or an independent identically distributed (i.i.d.) process. The order of the Volterra series is assumed to be known. The fitness function is defined as the difference between the calculated cumulant values and analytical equations in which the kernels and the input variances are considered. Simulation results and a comparative study for the proposed method and some existing techniques are given. They clearly show that the RCGA identification method performs better in terms of precision, time of convergence and simplicity of programming.

Détail

Keywords: blind nonlinear identification, Volterra series, higher order cumulants, real-coded genetic algorithm
Mots-clés : szereg Volterry, kumulanta wyższego rzędu, algorytm genetyczny kodowania rzeczywistego

Bibliographie

[1] Annaswamy, A.S. and Yu, H. (1996). Adaptive neural networks: A new approach to parameter estimation, IEEE Transactions on Neural Networks 7(4): 907–918.

[2] Attar, P.J. and Dowell, E.H. (2005). A reduced order system ID approach to the modelling of nonlinear structural behavior in aeroelasticity, Journal of Fluids and Structures 21(5–7): 531–542.

[3] Cherif, I., Abid, S., Fnaiech, F. and Favier, G. (2004). Volterra kernels identification using higher order moments for different input signals, IEEE-ISCCSP, Hammamet, Tunisia, pp. 845–848.

[4] Cherif, I., Abid, S. and Fnaiech, F. (2005). Blind identification of quadratic systems under i.i.d. excitation using genetic algorithms, 8th International Symposium on Signal Processing and its Applications ISSPA’2005, Sydney, Australia, pp. 463–466.

[5] Cherif I., Abid S., and Fnaiech, F. (2007). Blind nonlinear system identification under Gaussian and/or i.i.d. excitation using genetic algorithms, IEEE International Conference on Signal Processing and Communication ICSPC 2007, Dubai, UAE, pp. 644–647.

[6] Cherif, I., Abid, S., Fnaiech, F. (2012). Nonlinear blind identification with three dimensional tensor analysis, Mathematical Problems in Engineering 2012, Article ID: 284815.

[7] Glentis, G.O.A., Koukoulas, P. and Kalouptsidis, N. (1999). Efficient algorithms for Volterra system identification, IEEE Transactions on Signal Processing 47(11): 3042–3057.

[8] Greblicki, W. (2001). Recursive identification of Wiener systems, International Journal of Applied Mathematics and Computer Science 11(4): 977–991.

[9] Herrera, F., Lozano, M., and Verdegay, J.L. (1998). Tackling real-coded genetic algorithms: Operators and tools for behavioural analysis, Artificial Intelligence Review 12(4): 265–319.

[10] Koukoulas, P. and Kalouptsidis, N. (1996). On blind identification of quadratic systems, EUSIPCO’96, Trieste, Italy, pp. 1929–1932.

[11] Koukoulas, P. and Kalouptsidis, N. (1997). Third order system identification, International Conference on Acoustic Speech and Signal Processing: ICASSP’97, Munich, Germany, pp. 2405–2408.

[12] Koukoulas, P. and Kalouptsidis, N. (2000). Second order Volterra system identification, IEEE Transactions on Signal Processing 48(12): 3574–3577.

[13] Kalouptsidis, N. and Koukoulas, P. (2005). Blind identification of Volterra–Hammerstein systems, IEEE Transactions on Signal Processing 53(8): 2777–2787.

[14] Mathlouthi H., Abderrahim K., Msahli F. and Gerard F. (2009). Crosscumulants based approaches for the structure identification of Volterra models, International Journal of Automation and Computing 6(4): 420–430.

[15] Mendel J.M. (1991). Tutorial on higher order statistics (spectra) in signal processing and system theory: Theoretical results and some applications, Proceedings of the IEEE 79(3): 278–305.

[16] Ozertem, U. and Erdogmus, D. (2009). Second-order Volterra system identification with noisy input-output measurements, IEEE Signal Processing Letters 16(1): 18–21.

[17] Orjuela, R., Marx, B., Ragot, J. and Maquin D. (2013). Nonlinear system identification using heterogeneous multiple models, International Journal of Applied Mathematics and Computer Science 23(1): 103–115, DOI: 10.2478/amcs-2013-0009.

[18] Phan, M.Q., Longman, R.W., Lee, S.C. and Lee, J.-W. (2003). System identification from multiple-trial data corrupted by non-repeating periodic disturbances, International Journal of Applied Mathematics and Computer Science 13(2): 185–192.

[19] Stathaki, T. and Scohyers, A. (1997). A constrained optimisation approach to the blind estimation of Volterra kernels, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP-97, Munich, Germany, pp. 2373–2376.

[20] Stoica, P. and Soderstorm, T. (1982). Instrumental variable methods for identification of Hammerstein systems, International Journal of Control 35(3): 459–476.

[21] Tan, H.Z. and Chow, T.W.S. (2000). Blind identification of quadratic nonlinear models using neural networks with higher order cumulants, IEEE Transactions on Industrial Electronics 47(3): 687–696.

[22] Tseng, C.H. and Powers, E.J. (1995). Identification of cubic systems using higher order moments of i.i.d. signals, IEEE Transactions on Signal Processing 43(7): 1733–1735.

[23] Tsoulkas, V., Koukoulas, P. and Kalouptsidis, N. (2001). Identification of input-output bilinear system using cumulants, IEEE Transactions on Signal Processing 49(11): 2753–2761.

[24] Vasconcelos, J.A., Ramirez, J.A., Takahashi, R.H.C. and Saldanha, R.R. (2001). Improvement in genetic algorithms, IEEE Transactions on Magnetics 37(5): 3414–3417.

[25] Zhang, S. and Constantinides, A.G. (1992). Lagrange programming neural networks, IEEE Transactions on Circuits and Systems: Analog and Digital Signal Processing 39(7): 441–452.