An iterative Kalman smoother/least-squares algorithm for the identification of delta-ARX models

Additive measurement noise on the output signal is a significant problem in the δ-domain and disrupts parameter estimation of auto-regressive exogenous (ARX) models. This article deals with the identification of δ-domain linear time-invariant models of ARX structure (i.e. driven by known input signals and additive process noise) by using an iterative identification scheme, where the output is also corrupted by additive measurement noise. The identification proceeds by mapping the ARX model into a canonical state-space framework, where the states are the measurement noise-free values of the underlying variables. A consequence of this mapping is that the original parameter estimation task becomes one of both a state and parameter estimation problem. The algorithm steps between state estimation using a Kalman smoother and parameter estimation using least squares. This approach is advantageous as it avoids directly differencing the noise-corrupted ‘raw’ signals for use in the estimation phase and uses different techniques to the common parametric low-pass filters in the literature. Results of the algorithm applied to a simulation test problem as well as a real-world problem are given, and show that the algorithm converges quite rapidly and with accurate results.

[1]  Paul McKenna.  Delta operator:modelling forecasting and control , 1997 .

[2]  Graham C. Goodwin,et al.  On sampled-data models for nonlinear systems , 2005, IEEE Transactions on Automatic Control.

[3]  Brett Ninness,et al.  Robust maximum-likelihood estimation of multivariable dynamic systems , 2005, Autom..

[4]  Heinz Unbehauen,et al.  Continuous-time approaches to system identification - A survey , 1990, Autom..

[5]  Graham C. Goodwin,et al.  Rapprochement between continuous and discrete model reference adaptive control , 1986, Autom..

[6]  Petre Stoica,et al.  System identification from noisy measurements by using instrumental variables and subspace fitting , 1996 .

[7]  Alan V. Oppenheim,et al.  All-pole modeling of degraded speech , 1978 .

[8]  Michel Verhaegen,et al.  Linear and Non-linear System Identification Using Separable Least-Squares , 1997, Eur. J. Control.

[9]  H. Howard Fan,et al.  High Speed Adaptive Signal Progressing Using the Delta Operator, , 2001, Digit. Signal Process..

[10]  Bengt Carlsson,et al.  Estimation of continuous-time AR process parameters from discrete-time data , 1999, IEEE Trans. Signal Process..

[11]  J. S. Gibson,et al.  Least-squares estimation of input/output models for distributed linear systems in the presence of noise , 2000, Autom..

[12]  Amin Khodabakhshian,et al.  A New Identification Model for Power System Transfer Functions , 2003 .

[13]  G. Goodwin,et al.  High-speed digital signal processing and control , 1992, Proc. IEEE.

[14]  L. E. Scales,et al.  Introduction to Non-Linear Optimization , 1985 .

[15]  H. Unbehauen,et al.  Identification of continuous-time systems , 1991 .

[16]  Masayoshi Tomizuka,et al.  Parameter identification with derivative shift operator parametrization , 1999, Autom..

[17]  Wei Xing Zheng,et al.  Least-squares parameter estimation of linear systems with noisy input-output data , 2006, Int. J. Syst. Sci..

[18]  G. Goodwin,et al.  Improved finite word length characteristics in digital control using delta operators , 1986 .

[19]  V. Kadirkamanathan,et al.  Modelling and identification of non-linear deterministic systems in the delta-domain , 2007, Autom..

[20]  Torsten Söderström,et al.  Continuous-time AR process parameter estimation in presence of additive white noise , 1999, IEEE Trans. Signal Process..

[21]  Lennart Ljung,et al.  Initialisation aspects for subspace and output-error identification methods , 2003, 2003 European Control Conference (ECC).

[22]  Morten Bach Lauritsen,et al.  Delta-Domain Predictive Control and Identification for Control , 1997 .

[23]  Niels Kjølstad Poulsen,et al.  A state-space approach to the emulator-based GPC design , 1996 .

[24]  Peter C. Young,et al.  A Direct Approach to the Identification and Estimation of Continuous-time Systems from Discrete-time Data Based on Fixed Interval Smoothing , 1993 .

[25]  Charles P. Neuman,et al.  Properties of the delta operator model of dynamic physical systems , 1993, IEEE Trans. Syst. Man Cybern..

[26]  Graham C. Goodwin,et al.  Digital control and estimation : a unified approach , 1990 .

[27]  Sean R. Anderson,et al.  Predictive control of fast-sampled systems using the delta-operator , 2009, Int. J. Syst. Sci..

[28]  L. Jetto,et al.  Accurate derivative estimation from noisy data: a state-space approach , 1989 .

[29]  Arie Yeredor,et al.  The extended least squares criterion: minimization algorithms and applications , 2001, IEEE Trans. Signal Process..

[30]  Emmanuel G. Collins,et al.  A delta operator approach to discrete-time H ∞ control , 1999 .

[31]  R. Shumway,et al.  AN APPROACH TO TIME SERIES SMOOTHING AND FORECASTING USING THE EM ALGORITHM , 1982 .

[32]  Gene H. Golub,et al.  The differentiation of pseudo-inverses and non-linear least squares problems whose variables separate , 1972, Milestones in Matrix Computation.

[33]  T. Söderström,et al.  Least squares parameter estimation of continuous-time ARX models from discrete-time data , 1997, IEEE Trans. Autom. Control..

[34]  Heinz Unbehauen,et al.  Continuous-time Approaches to System Identification , 1988 .