Linear-wavelet networks

Galvao, R. K. H. ; Becerra, V. M. ; Calado, J. M. F. ; Silva, P. M.

International Journal of Applied Mathematics and Computer Science, Tome 14 (2004) no. 2, pp. 221-232.

Résumé

This paper proposes a nonlinear regression structure comprising a wavelet network and a linear term. The introduction of the linear term is aimed at providing a more parsimonious interpolation in high-dimensional spaces when the modelling samples are sparse. A constructive procedure for building such structures, termed linear-wavelet networks, is described. For illustration, the proposed procedure is employed in the framework of dynamic system identification. In an example involving a simulated fermentation process, it is shown that a linear-wavelet network yields a smaller approximation error when compared with a wavelet network with the same number of regressors. The proposed technique is also applied to the identification of a pressure plant from experimental data. In this case, the results show that the introduction of wavelets considerably improves the prediction ability of a linear model. Standard errors on the estimated model coefficients are also calculated to assess the numerical conditioning of the identification process.

Détail

Keywords: wavelet networks, nonlinear models, regression analysis, system identification
Mots-clés : sieć falkowa, model nieliniowy, analiza regresji, identyfikacja systemu

Bibliographie

[1] Aborhey S. and Williamson D. (1978): State and parameter estimation of microbial growth processes. — Automatica, Vol. 14, No. 5, pp. 493–498.

[2] Benveniste A., Juditsky A., Delyon B., Zhang Q. and Glorennec P.Y. (1994): Wavelets in identification.— Proc. 10th IFAC Symp. Syst. Identification, Copenhagen, pp. 27–48.

[3] Cannon M. and Slotine J.-J.E. (1995): Space-frequency localized basis function networks for nonlinear system estimation and control.—Neurocomput., Vol. 9, No. 3, pp. 293–342.

[4] D’Ans G., Gottlieb D. and Kokotovic P. (1972): Optimal control of bacterial growth.—Automatica, Vol. 8, No. 6, pp. 729–736.

[5] Daubechies I. (1992): Ten Lectures on Wavelets. — Philadelphia: SIAM.

[6] Draper N.R. and Smith H. (1981): Applied Regression Analysis, 2nd Ed. — New York: Wiley.

[7] Ezekiel M. and Fox K.A. (1959): Methods of Correlation and Regression Analysis, 3rd Ed. — New York: Wiley.

[8] Galvão R.K.H. and Becerra V.M. (2002): Linear-wavelet models applied to the identification of a two-link manipulator. —Proc. 21st IASTED Int. Conf. Modelling, Identification and Control, Innsbruck, pp. 479–484.

[9] Galvão R.K.H., Yoneyama T. and Rabello T.N. (1999): Signal representation by adaptive biased wavelet expansions. — Digital Signal Process., Vol. 9, No. 4, pp. 225–240.

[10] Haykin S.S. (1998): Neural Networks: A Comprehensive Foundation. — Upper Saddle River: Prentice-Hall.

[11] Jang J.-S. R. and Sun C.-T. (1995): Neuro-fuzzy modelling and control. — Proc. IEEE, Vol. 83, No. 3, pp. 378–406.

[12] Kan K.-C. and Wong K.-W. (1998): Self-construction algorithm for synthesis of wavelet networks.—Electronic Lett., Vol. 34, No. 20, pp. 1953–1955.

[13] Lawson C.L. and Hanson R.J. (1974): Solving Least Squares Problems. — Englewood Cliffs: Prentice-Hall.

[14] Li K.C. (1986): Asymptotic optimality of cl and generalized cross-validation in ridge regression and application to the spline smoothing. — Ann. Statist., Vol. 14, No. 3, pp. 1101–1112.

[15] Liu G.P., Billings S.A. and Kadirkamanathan V. (2000): Nonlinear system identification using wavelet networks. — Int. J. Syst. Sci., Vol. 31, No. 12, pp. 1531–1541.

[16] Ljung L. (1999): System Identification: Theory for the User. — Upper Saddle River: Prentice-Hall.

[17] Naes T. and Mevik B.H. (2001): Understanding the collinearity problem in regression and discriminant analysis. — J. Chemometr., Vol. 15, No. 4, pp. 413–426.

[18] Naradaya E. (1964): On estimating regression. — Theory Prob. Applicns., Vol. 9, pp. 141–142.

[19] Narendra K.S. and Parthasarathy K. (1990): Identification and control of dynamical systems using neural networks. — IEEE Trans. Neural Netw., Vol. 1, No. 1, pp. 4–27.

[20] Poggio T. and Girosi F. (1990): Networks for approximation and learning.—Proc. IEEE, Vol. 78, No. 9, pp. 1481–1497.

[21] Rissanen J. (1978): Modeling by shortest data description. — Automatica, Vol. 14, No. 5, pp. 465–471.

[22] Rugh W.J. (1981): Nonlinear Systems Theory. The Volterra/Wiener Approach. — Baltimore: Johns Hopkins University Press.

[23] Schumaker L.L. (1981): Spline Functions: Basic Theory. — Chichester: Wiley.

[24] Souza Jr. C., Hemerly E.M. and Galvão R.K.H. (2002): Adaptive control for mobile robot using wavelet network. — IEEE Trans. Syst. Man Cybern., Part B, Vol. 32, No. 4, pp. 493–504.

[25] Takagi T. and Sugeno M. (1985): Fuzzy identification of systems and its applications to modelling and control. — IEEE Trans. Syst. Man Cybern., Vol. 15, No. 1, pp. 116-132.

[26] Watson G.S. (1964): Smooth regression analysis. — Sankhya, Ser. A, Vol. 26, No. 4, pp. 359–372.

[27] Zhang J.,Walter G.G., Miao Y. and Lee W.N.W. (1995): Wavelet neural networks for function learning.—IEEE Trans. Signal Process., Vol. 43, No. 6, pp. 1485–1496.

[28] Zhang Q. (1997): Using wavelet network in nonparametric estimation. — IEEE Trans. Neural Netw., Vol. 8, No. 2, pp. 227–236.

[29] Zhang Q. and Benveniste A. (1992): Wavelet networks.—IEEE Trans. Neural Netw., Vol. 3, No. 6, pp. 889–898.