Evolutionary parametric identification of dynamic systems

This paper presents a new method for the estimation of Single Input-Single Output, Autoregressive Moving Average with exogenous inputs (ARMAX) models, by means of Prediction-Error Methods (PEM). It’s main feature lies in the use of a hybrid optimization algorithm, capable of giving superior performance in PEM. The new method turns to be more flexible than conventional prediction-error techniques, since no initial “guess” for the parameter vector is required, while stability is guaranteed. For the practical implementation of the new method, a testing apparatus that consists of a flexible robotic arm driven by a servomotor has been used, and a corresponding input-output data set has been acquired. Evolutionäre parametrische Erkinnung dynamischer SystemeZusammenfassung In diese Arbeit wird eine neue Methode für die Ermittlung des Verhaltens von sog. ARMAX-Modellen mit Hilfe der Fehler-Vorhersage-Methode (FVM) vorgestellt. Eine wesentliche Besonderheit besteht in der Verwendung eines neuen Hybrid Optimierungsalgorithmus, der die Leistung der FVM steigert. Die neue Methode ist flexibler als bisherige Vorgehensweisen, da eine Startwertvorgabe für die gesuchten Parameter nicht erforderlich ist. Der Erfolg der Methode wird exemplarisch an einer Testapparatur mit einem flexiblen Roboterarm dargestellt.

[1]  David J. Murray-Smith,et al.  Nonlinear model structure identification using genetic programming , 1998 .

[2]  Ajay K. Ray,et al.  Multi-objective optimization of industrial hydrogen plants , 2001 .

[3]  Nicolas Monmarché,et al.  On how Pachycondyla apicalis ants suggest a new search algorithm , 2000, Future Gener. Comput. Syst..

[4]  Spilios D. Fassois,et al.  PARAMETRIC TIME-DOMAIN METHODS FOR THE IDENTIFICATION OF VIBRATING STRUCTURES—A CRITICAL COMPARISON AND ASSESSMENT , 2001 .

[5]  R. Fletcher.  Practical Methods of Optimization , 1988 .

[6]  Marco Dorigo,et al.  Ant algorithms and stigmergy , 2000, Future Gener. Comput. Syst..

[7]  R. Ben Mrad,et al.  ON-BOARD PREDICTION OF POWER CONSUMPTION IN AUTOMOBILE ACTIVE SUSPENSION SYSTEMS. II. VALIDATION AND PERFORMANCE EVALUATION , 1996 .

[8]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[9]  Peter J. Fleming,et al.  Evolutionary algorithms in control systems engineering: a survey , 2002 .

[10]  R. Ben Mrad,et al.  ON-BOARD PREDICTION OF POWER CONSUMPTION IN AUTOMOBILE ACTIVE SUSPENSION SYSTEMS—I: PREDICTOR DESIGN ISSUES , 1996 .

[11]  Kay Chen Tan,et al.  Grey-box model identification via evolutionary computing , 2002 .

[12]  Thomas Bäck,et al.  Evolutionary Algorithms in Theory and Practice , 1996 .

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

[14]  Stephen A. Billings,et al.  Structure detection for nonlinear rational models using genetic algorithms , 1998, Int. J. Syst. Sci..

[15]  Alan V. Oppenheim,et al.  Discrete-Time Signal Pro-cessing , 1989 .

[16]  Guy Albert Dumont,et al.  System identification and control using genetic algorithms , 1992, IEEE Trans. Syst. Man Cybern..

[17]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1971 .

[18]  Petre Stoica,et al.  Decentralized Control , 2018, The Control Systems Handbook.