Study of the Variance of Parametric Estimates of the Best Linear Approximation of Nonlinear Systems

Abstract-This paper analyzes the variance of the estimated parametric best linear approximation (BLA) Ĝ<sub>BLA</sub>(q,θ) of a nonlinear system that is driven by random excitations. The estimated model Ĝ<sub>BLA</sub>(q; θ) varies not only due to the disturbing measurement and process noise but also over different realizations of the random excitation because the nonlinear distortions depend on the input realization. For the nonparametric frequency response function (FRF) estimate, it has been shown that the variance expression is still valid in the presence of nonlinear distortions, and the same formulas can be used in the linear as in the nonlinear case. This result does not hold for the variance σ(Ĝ<sub>BLA</sub>(q,θ))<sup>2</sup> on the parametric estimate Ĝ<sub>BLA</sub>(q,θ). It is shown in this paper that it is still possible to upperbound the variance σ(Ĝ<sub>BLA</sub>(q,θ))<sup>2</sup> using linear expression by introducing an additional scaling factor that depends upon the maximal degree of nonlinearity.

[1]  Lennart Ljung,et al.  Linear approximations of nonlinear FIR systems for separable input processes , 2005, Autom..

[2]  M. Enqvist.  Linear models of nonlinear systems , 2005 .

[3]  Pertti M. Mäkilä,et al.  LTI approximation of nonlinear systems via signal distribution theory , 2006, Autom..

[4]  Lennart Ljung,et al.  Model Error Modeling and Control Design , 2000 .

[5]  Rik Pintelon,et al.  System Identification: A Frequency Domain Approach , 2012 .

[6]  Michel Gevers,et al.  Quantification of the variance of estimated transfer functions in the presence of undermodeling , 2004, IEEE Transactions on Automatic Control.

[7]  Gerd Vandersteen,et al.  Robustness Issues of the Best Linear Approximation of a Nonlinear System , 2009, IEEE Transactions on Instrumentation and Measurement.

[8]  Julian J. Bussgang,et al.  Crosscorrelation functions of amplitude-distorted gaussian signals , 1952 .

[9]  M. Schetzen.  The Volterra and Wiener Theories of Nonlinear Systems , 1980 .

[10]  Yves Rolain,et al.  Uncertainty of transfer function modelling using prior estimated noise models , 2003, Autom..

[11]  Pertti M. Mäkilä,et al.  On optimal LTI approximation of nonlinear systems , 2004, IEEE Trans. Autom. Control..

[12]  J. Schoukens,et al.  Parametric and nonparametric identification of linear systems in the presence of nonlinear distortions-a frequency domain approach , 1998, IEEE Trans. Autom. Control..

[13]  Julius S. Bendat,et al.  Engineering Applications of Correlation and Spectral Analysis , 1980 .

[14]  J. Schoukens,et al.  Estimation of nonparametric noise and FRF models for multivariable systems—Part I: Theory , 2010 .

[15]  Jonathan R. Partington,et al.  Least-squares LTI approximation of nonlinear systems and quasistationarity analysis , 2004, Autom..

[16]  Rik Pintelon,et al.  Nonlinear Induced Variance of Frequency Response Function Measurements , 2010, IEEE Transactions on Instrumentation and Measurement.

[17]  Alex Simpkins,et al.  System Identification: Theory for the User, 2nd Edition (Ljung, L.; 1999) [On the Shelf] , 2012, IEEE Robotics & Automation Magazine.

[18]  Rik Pintelon,et al.  Variance analysis of frequency response function measurements using periodic excitations , 2004, IMTC 2004.

[19]  Lennart Ljung,et al.  Model Validation and Model Error Modeling , 1999 .

[20]  E. Brigham,et al.  The fast Fourier transform , 2016, IEEE Spectrum.

[21]  Lennart Ljung,et al.  Comparing different approaches to model error modeling in robust identification , 2002, Autom..

[22]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[23]  J. Schoukens,et al.  Robustness Issues of the Equivalent Linear Representation of a Nonlinear System , 2008, 2008 IEEE Instrumentation and Measurement Technology Conference.