Abstract
This chapter summarizes the most important issues of the state-of-the-art linear system identification. It introduces the standard terminology and explains its origin as it can be confusing at first. Dynamic models are structured into (i) time series, (ii) model with output feedback, and (iii) model without output feedback. The latter model class typically is underrepresented and underappreciated in most of the literature, in the opinion of the author. Therefore, it is treated more extensively here and also in other chapters of this book. Care is taken to explain the decisive difference between equation (or one-step prediction) error and output (or simulation) error and the consequences associated with their properties. Advanced issues like order determination, multivariate systems, and closed-loop identification are briefly discussed as well.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In a considerable part of the literature and in older publications in particular, the ARX model is called an “ARMA” model to express the fact that both a numerator and a denominator polynomial exist. However, as discussed above, this book follows the current standard terminology established in [356], where ARMA stands for the time series model in (18.9).
- 2.
In some cases, the signal may not depend on time but rather on space (e.g., in geology) or some other quantity.
- 3.
Often ELS denotes the recursive version of this algorithm. Here, for the sake of clarity, the recursive algorithm is named RELS; see Sect. 18.8.3.
- 4.
Often for unstable processes, a simple stabilizing controller is designed, and subsequently, the resulting stable closed-loop system is identified to design a second, more advanced controller for the inner closed-loop system.
- 5.
Filtering requires the choice of a filter bandwidth, and with this choice, certain frequency ranges are emphasized in the model fit; see Sect. 18.7.4. Note that a reasonable choice of the filter bandwidth also requires prior knowledge of the process dynamics.
- 6.
Sometimes denoted only as extended least squares (ELS) [356].
References
Box, G.E.P., Jenkins, G.M.: Times Series Analysis, Forecasting and Control. Holden-Day, San Francisco (1970)
Brockwell, P.J., Davis, R.A.: Introduction to Time Series and Forecasting. Springer, New York (1996)
Chen, T., Ljung, L.: Implementation of algorithms for tuning parameters in regularized least squares problems in system identification. Automatica 49(7), 2213–2220 (2013)
Eykhoff, P.: System Identification. John Wiley & Sons, London (1974)
Gevers, M., Ljung, L.: Optimal experiment design with respect to the intended model application. Automatica 22(5), 543–554 (1986)
Haykin, S.: Adaptive Filter Theory. Prentice Hall, Oxford (1991)
Heuberger, P.S.C., Van den Hof, P.M.J., Bosgra, O.H.: Modelling linear dynamical systems through generalized orthonormal basis functions. In: IFAC World Congress, vol. 5, pp. 19–22, Sydney, Australia (1993)
Heuberger, P.S.C., Van den Hof, P.M.J., Bosgra, O.H.: A generalized orthonormal basis for linear dynamical systems. IEEE Trans. Autom. Control 40(3), 451–465 (1995)
Hjalmarsson, H., Gevers, M., de Bruyne, F.: For model-based control design, closed-loop identification gives better performance. Automatica 32(12), 1659–1673 (1996)
Isermann, R.: Required accuracy of mathematical models of linear time invariant controlled elements. Automatica 7, 333–341 (1971)
Isermann, R.: Digitale Regelsysteme – Band 1, 2. ed. Springer, Berlin (1987)
Isermann, R.: Identifikation dynamischer Syteme – Band 1, 2. ed. Springer, Berlin (1992)
Isermann, R.: Identifikation dynamischer Syteme – Band 2, 2. ed. Springer, Berlin (1992)
Isermann, R., Baur, U., Bamberger, W., Kneppo, P., Siebert, H.: Comparision of six on-line identification and parameter estimation methods. Automatica 10, 81–103 (1974)
Isermann, R., Lachmann, K.-H., Matko, D.: Adaptive Control Systems. Prentice Hall, New York (1992)
Johansen, T.A.: Operating Regime Based Process Modeling and Identification. Ph.D. thesis, Dept. of Engineering Cybernetics, Norwegian Institute of Technology, Trondheim, Norway (1994)
Johansson, R.: System Modeling & Identification. Prentice Hall, Englewood Cliffs (1993)
Lim, R.K., Phan, M.Q.: Identification of a multistep-ahead observer and its application to predictive control. J. Guid. Control. Dyn. 20(6), 1200–1206 (1997)
Lindskog, P.: Methods, Algorithms and Tools for System Identification Based on Prior Knowledge. Ph.D. thesis, Linköping University, Linköping, Sweden (1996)
Ljung, L.: System Identification: Theory for the User, 2nd edn. Prentice Hall, Englewood Cliffs (1999)
Ljung, L.: System identification toolbox for MATLAB user’s guide. The Matlab User’s Guide (1988)
Marconato, A., Schoukens, M.: Tuning the hyperparameters of the filter-based regularization method for impulse response estimation. IFAC-PapersOnLine 50(1), 12841–12846 (2017)
Marconato, A., Schoukens, M., Schoukens, J.: Filter-based regularisation for impulse response modelling. IET Control Theory Appl. (2016)
McKelvey, T.: Identification of State-Space Models from Time and Frequency Data. Ph.D. thesis, Linköping University, Linköping, Sweden (1995)
Münker, T., Belz, J., Nelles, O.: Improved incorporation of prior knowledge for regularized FIR model identification. In: 2018 Annual American Control Conference (ACC), pp. 1090–1095, Milwaukee, WI. IEEE (2018)
Münker, T., Peter, T., Nelles, O.: Gray-box identification with regularized FIR models. at-Automatisierungstechnik 66(9), 704–713 (2018)
Pillonetto, G., Dinuzzo, F., Chen, T., De Nicolao, G., Ljung, L.: Kernel methods in system identification, machine learning and function estimation: a survey. Automatica 50(3), 657–682 (2014)
Saridis, G.N.: Comparision of six on-line identification algorithms. Automatica 10, 69–79 (1974)
Schenker, B., Agarwal, M.: Long-range prediction for poorly-known systems. Int. J. Control 62(1), 227–238 (1995)
Schenker, B., Agarwal, M.: Dynamic modelling using neural networks. Int. J. Syst. Sci. 28(12), 1285–1298 (1997)
Schenker, B.G.E.: Prediction and Control Using Feedback Neural Networks and Partial Models. Ph.D. thesis, Swiss Federal Institute of Technology Zürich, Zürich, Switzerland (1996)
Söderström, T., Stoica, P.: System Identification. Prentice Hall, New York (1989)
Van den Hof, P.M.J.: Closed-loop issues in system identification. In: IFAC Symposium on System Identification, pp. 1651–1664, Fukuoka, Japan (1997)
Van den Hof, P.M.J., Heuberger, P.S.C., Bokor, J.: System identification with generalized orthonormal basis functions. Automatica 31(12), 1821–1834 (1995)
Van Overschee, P., De Moor, B.: N4SID: subspace algorithms for the identification of combined deterministic-stochastic systems. Automatica 30(1), 75–93 (1994)
Verhagen, M.: Identification of the deterministic part of MIMO state space models given in innovations form from input-output data. Automatica 30(1), 61–74 (1993)
Viberg, M.: Subspace-based methods for the identification of linear time-invariant systems. Automatica 31(12), 1835–1851 (1995)
Wahlberg, B.: System identification using Laguerre models. IEEE Trans. Autom. Control 36(5), 551–562 (1991)
Wahlberg, B.: System identification using Kautz models. IEEE Trans. Autom. Control 39(6), 1276–1282 (1994)
Zhan, J., Ishida, M.: The multi-step predictive control of nonlinear siso processes with a neural model predictive control (NMPC) method. Comput. Chem. Eng. 21(2), 201–210 (1997)
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Nelles, O. (2020). Linear Dynamic System Identification. In: Nonlinear System Identification. Springer, Cham. https://doi.org/10.1007/978-3-030-47439-3_18
Download citation
DOI: https://doi.org/10.1007/978-3-030-47439-3_18
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-47438-6
Online ISBN: 978-3-030-47439-3
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)