Abstract
This chapter addresses many fundamental issues arising when transitioning from nonlinear static to nonlinear dynamic models. Many aspects are very general in nature and independent of the specific model architecture. They are analyzed here. The two competing concepts of external and internal dynamics are contrasted. It is explained how equation and output errors are traditionally treated in the neural network terminology as series-parallel and parallel model structures. The additional difficulties feedback causes are discussed and how they are dealt with in learning as well. A large section discusses the much-neglected topic of suitable excitation signals for nonlinear dynamic processes and new ideas for their analysis. Finally, a new signal generator is proposed that is capable of generating very good excitation signals, which is much more flexible and in contrast to the conventional approach of postulating a specific parameterized signal type.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This analysis is carried out for the external dynamics approach because it allows us to gain some important insights about the desirable properties of the excitation signals. Although with the internal dynamics approach the one-step prediction function is not explicitly approximated, this analysis based on information content considerations is also valid for this class of approaches.
References
Baumann, W., Schaum, S., Roepke, K., Knaak, M.: Excitation signals for nonlinear dynamic modeling of combustion engines. In: Proceedings of the 17th World Congress, The International Federation of Automatic Control, Seoul, Korea (2008)
Billings, S.A., Voon, W.S.F.: Correlation based validity tests for nonlinear models. Int. J. Control 44(1), 235–244 (1986)
Billings, S.A., Zhu, Q.M.: Nonlinear model validation using correlation tests. Int. J. Control 60(6), 1107–1120 (1994)
Box, G.E.P., Meyer, R.D.: An analysis for unreplicated fractional factorials. Technometrics 28(1), 11–18 (1986)
Chen, S., Billings, S.A.: Representations of nonlinear systems: the NARMAX model. Int. J. Control 49(3), 1013–1032 (1983)
Chikkula, Y., Lee, J.H.: Input sequence design for parametric identification of nonlinear systems. In: American Control Conference (ACC), vol. 5, pp. 3037–3041, Albuquerque, USA (1997)
Deflorian, M., et al.: Versuchsplanung und Methoden zur Identifikation zeitkontinuierlicher Zustandsraummodelle am Beispiel des Verbrennungsmotors. Ph.D. thesis, Technische Universität München (2011)
Dotsenko, V.I., Faradzhev, R.G., Chakartisvhili, G.S.: Properties of maximum length sequences with p-levels. Automatika i Telemechanika 8, 189–194 (1971)
Ebert, T., Fischer, T., Belz, J., Heinz, T.O., Kampmann, G., Nelles, O.: Extended deterministic local search algorithm for maximin Latin hypercube designs. In: 2015 IEEE Symposium Series on Computational Intelligence: IEEE Symposium on Computational Intelligence in Control and Automation (2015 IEEE CICA), Cape Town, South Africa (2015)
Eykhoff, P.: System Identification. John Wiley & Sons, London (1974)
Fischer, M., Nelles, O., Isermann, R.: Exploiting prior knowledge in fuzzy model identification of a heat exchanger. In: IFAC Symposium on Artificial Intelligence in Real-Time Control (AIRTC), pp. 445–450, Kuala Lumpur, Malaysia (1997)
Haber, R., Keviczky, L.: Identification of ‘linear’ systems having state-dependent parameters. Int. J. Syst. Sci. 16(7), 869–884 (1985)
Häck, M., Köhne, M.: Internal Model Control mit neuronalen Netzen zur Regelung eines Prozessanalysators. Automatisierungstechnik 45(1) (1997)
Hametner, C., Stadlbauer, M., Deregnaucourt, M., Jakubek, S., Winsel, T.: Optimal experiment design based on local model networks and multilayer perceptron networks. Eng. Appl. Artif. Intell. 26(1), 251–261 (2013)
Haykin, S.: Neural Networks. A Comprehensive Foundation. Macmillan, New York (1994)
He, X., Asada, H.: A new method for identifying orders of input-output models for nonlinear dynamic systems. In: American Control Conference (ACC), pp. 2520–2524, San Francisco, USA (1993)
Heinz, T.O., Nelles, O.: Efficient pole optimization of nonlinear Laguerre filter models. In: 2015 IEEE Symposium Series on Computational Intelligence, pp. 1–7. IEEE (2016)
Heinz, T.O., Nelles, O.: Vergleich von anregungssignalen für nichtlineare identifikationsaufgaben. In: Hoffman, F., Hüllermeier, E., Mikut, R. (eds.) Proceedings 26. Workshop Computational Intelligence, pp. 139–158. KIT Scientific Publishing (2016)
Heinz, T.O., Nelles, O.: Iterative excitation signal design for nonlinear dynamic black-box models. Proc. Comput. Sci. 112, 1054–1061 (2017)
Heinz, T.O., Nelles, O.: Excitation signal design for nonlinear dynamic systems with multiple inputs – a data distribution approach. at-Automatisierungstechnik 66(9), 714–724 (2018)
Heinz, T.O., Schillinger, M., Hartmann, B., Nelles, O.: Excitation signal design for nonlinear dynamic systems. In: Röpke, K., Gühmann, C. (eds.) International Calibration Conference – Automotive Data Analytics, Methods, DoE, pp. 191–208. expertVerlag (2017)
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)
Isermann, R.: Identifikation dynamischer Syteme – Band 1, 2. ed. Springer, Berlin (1992)
Isermann, R.: Identifikation dynamischer Syteme – Band 2, 2. ed. Springer, Berlin (1992)
Johnson, M.E., Moore, L.M., Ylvisaker, D.: Minimax and maximin distance designs. J. Stat. Plan. Inference 26(2), 131–148 (1990)
Leontaritis, I.J., Billings, S.A.: Input-output parametric models for nonlinear systems, part 1: deterministic nonlinear systems. Int. J. Control 41, 303–344 (1985)
Levin, A.U., Narendra, K.S.: Identification using feedforward networks. Neural Comput. 7(2), 349–357 (1995)
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., Sjöberg, J., Suykens, J.A.K., Schoukens, J.: Improved initialization for nonlinear state-space modeling. IEEE Trans. Instrum. Meas. 63(4), 972–980 (2014)
Montgomery, D.C.: Design and Analysis of Experiments. John Wiley & Sons, Hoboken (2008)
Morris, M.D., Mitchell, T.J.: Exploratory designs for computational experiments. J. Stat. Plan. Inference 43(3), 381–402 (1995)
Narendra, K.S., Parthasarathy, K.: Identification and control of dynamical systems using neural networks. IEEE Trans. Neural Netw. 1(1), 4–27 (1990)
Nelles, O.: On the identification with neural networks as series-parallel and parallel models. In: International Conference on Artificial Neural Networks (ICANN), pp. 255–260, Paris, France (1995)
Nelles, O.: Training neuronaler Netze als seriell-parallele oder parallele Modelle zur Identifikation nichtlinearer, dynamischer Systeme. In: 3. GI Fuzzy-Neuro-Systeme Workshop, pp. 333–339, Darmstadt, Germany (1995)
Nelles, O., Isermann, R.: A comparison between radial basis function networks and classical methods for identification of nonlinear dynamic systems. In: IFAC Symposium on Adaptive Systems in Control and Signal Processing (ASCSP), pp. 233–238, Budapest, Hungary (1995)
Nelles, O., Isermann, R.: Identification of nonlinear dynamic systems – classical methods versus radial basis function networks. In: American Control Conference (ACC), pp. 3786–3790, Seattle, USA (1995)
Paduart, J., Lauwers, L., Swevers, J., Smolders, K., Schoukens, J., Pintelon, R.: Identification of nonlinear systems using polynomial nonlinear state space models. Automatica 46(4), 647–656 (2010)
Pearlmutter, B.A.: Learning state space trajectories in recurrent neural networks. Neural Comput. 1(2), 263–269 (1989)
Pearlmutter, B.A.: Gradient calculations for dynamic recurrent neural networks: a survey. IEEE Trans. Neural Netw. 6(5), 1212–1228 (1995)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning internal representations by error propagation. In: Rumelhart, D.E., McClelland, J.L. (eds.) Parallel Distributed Processing: Explorations in the Mircostructure of Cognition, vol. 1, chapter 8. MIT Press, Cambridge (1986)
Schoukens, J., Vaes, M., Pintelon, R.: Linear system identification in a nonlinear setting: nonparametric analysis of the nonlinear distortions and their impact on the best linear approximation. IEEE Control Syst. 36(3), 38–69 (2016)
Sjöberg, J.: Non-Linear System Identification with Neural Networks. Ph.D. thesis, Linköping University, Linköping, Sweden (1995)
Sjöberg, J., Zhang, Q., Ljung, L., Benveniste, A., Delyon, B., Glorennec, P.-Y., Hjalmarsson, H., Juditsky, A.: Nonlinear black-box modeling in system identification: a unified overview. Automatica 31(12), 1691–1724 (1995)
Škrjanc, I.: Evolving fuzzy-model-based design of experiments with supervised hierarchical clustering. IEEE Trans. Fuzzy Syst. 23(4), 861–871 (2015)
Söderström, T., Stoica, P.: System Identification. Prentice Hall, New York (1989)
Tuis, L.: Anwendung von mehrwertigen pseudozufälligen Signalen zur Identifikation von nichtlinearen Regelungssystemen. Ph.D. thesis, Lehrstuhl für Mess- und Regelungstechnik, Abt. Maschinenbau, Ruhr-Universität Bochum, Bochum, Germany (1975)
Ullrich, T.: Untersuchungen zur effizienten interpolierenden Speicherung von nichtlinearen Prozeßmodellen und Vorsteuerungsstrategien: Methodik und Anwendungen in der Automobilelektronik. Automatisierungstechnik series. Shaker Verlag, Aachen. Ph.D. Thesis, TU Darmstadt (1998)
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)
Weigend, A.S., Gershenfeld, N.A.: Time series prediction: forecasting the future and understanding the past. Addison-Wesley, Reading (1994)
Williams, R.J.: Adaptive state representation and estimation using recurrent connectionist. In: Miller, W.T., Sutton, R.S., Werbos, P.J. (eds.) Neural Networks for Control, pp. 97–114. MIT Press, Cambridge (1990)
Williams, R.J., Zipser, D.: A learning algorithm for continually running fully recurrent neural networks. Neural Comput. 1(2), 270–280 (1989)
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). Nonlinear Dynamic System Identification. In: Nonlinear System Identification. Springer, Cham. https://doi.org/10.1007/978-3-030-47439-3_19
Download citation
DOI: https://doi.org/10.1007/978-3-030-47439-3_19
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)