A Novel Dual-Error Approach to System Identification
Single error system identification techniques are widely used to estimate the parameters of dynamic mathematical models that are needed in a range of industrial applications. A novel Dual-Error system identification technique is proposed. It is based on a modification of the traditional single-error methods and shown to offer better accuracy for the estimation of model parameters. The benefits of the proposed method are demonstrated by a comparison with traditional methods when applied to both a simulated system and a DC motor.
KeywordsSystem identification Dynamic system identification Parameter estimation Generalised error single error Dual-error
The author would like to thank Professor M. Braae for his help, whose patience, guidance and motivation made this project possible.
- 1.G. P. Rao and H. Unbehauen, “Identification of continuous-time systems,” IEE Proceedings - Control Theory, vol. 153, no. 2, pp. 185–220, 2006.Google Scholar
- 2.L. Ljung, System Identification: Theory for the User. Englewood Cliffs, New Jersey: Prentice Hall, 1987.Google Scholar
- 3.T. Söderström and P. Stoica, System identification. Hertfordshire: Prentice Hall International, 1989.Google Scholar
- 4.M. Bodson and S. Sastry, “Input Error versus Output Error Model Reference Adaptive Control,” in American Control Conference, 1987, pp. 224–229.Google Scholar
- 5.Y. Tomita, A. A. H. Damen, and P. M. J. Van Den Hof, “Equation error versus output error methods,” ERGONOMICS, vol. 35, no. 5/6, pp. 551–564, 1992.Google Scholar
- 6.Y. D. Landau, Adaptive control: the model reference approach, First Edit. New York: Marcel Dekker, 1979.Google Scholar
- 7.H. C. So, Y. T. Chan, K. C. Ho, and K. W. Chan, “UNBIASED EQUATION-ERROR APPROACH FOR EFFICIENT IIR SYSTEM IDENTIFICATION,” in Proc. 2004 European Signal Processing Conference, 2004, pp. 1907–1910.Google Scholar