Identification of a linear model for nonlinear systems

作者: Jiong Tang , Rajamani Doraiswami , Chris P. Diduch

DOI: 10.1109/ICCA.2009.5410381

关键词:

摘要: It is shown that the output of a certain class nonlinear dynamic systems can match arbitrarily close linear system if spectral content probing input same as system. The with cascade or feedback combination static elements and considered. If nonlinearity odd symmetric, signal periodic, persistently exciting only harmonics then it an between may be achieved. proposed method differs from traditional approximation model in captures behavior over larger region operating point. scheme verified on simulated systems, tested physical system, finds application identification for control design, fault diagnosis, analysis

参考文章(4)
K.J. Hunt, D. Sbarbaro, R. Żbikowski, P.J. Gawthrop, Neural networks for control systems: a survey Automatica. ,vol. 28, pp. 1083- 1112 ,(1992) , 10.1016/0005-1098(92)90053-I
R. Doraiswami, Two-stage identification with applications to control, feature extraction, and spectral estimation IEE Proceedings - Control Theory and Applications. ,vol. 152, pp. 379- 386 ,(2005) , 10.1049/IP-CTA:20041122
G.J.W. Mallory, R. Doraiswami, A filter for on-line estimation of spectral content IEEE Transactions on Instrumentation and Measurement. ,vol. 48, pp. 1047- 1055 ,(1999) , 10.1109/19.816112
W.A. Docter, C. Georgakis, Identification of reduced order average linear models from nonlinear dynamic simulations american control conference. ,vol. 5, pp. 3047- 3052 ,(1997) , 10.1109/ACC.1997.612017