Quantifying the accuracy of Hammerstein model estimation

被引:46
作者
Ninness, B [1 ]
Gibson, S [1 ]
机构
[1] Univ Newcastle, Dept Elect & Comp Engn, Newcastle, NSW 2308, Australia
基金
澳大利亚研究理事会;
关键词
Hammerstein models; system identification; parameter estimation; non-linear systems;
D O I
10.1016/S0005-1098(02)00101-2
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper investigates the accuracy of the linear component that forms part of an overall Hammerstein model-structure estimate, and a key finding is that the process of estimating the non-linear element can have a strong effect on the associated estimate of the linear dynamics. Furthermore, this effect is not explained simply by way of considering how the input spectrum is changed by the non-linearity. Instead, it arises that the linear model-estimate variability may be dominated by a term that depends on the frequency response of the linear system itself. Amongst other things, the main results derived here have experiment design implications for Hammerstein system estimation. (C) 2002 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:2037 / 2051
页数:15
相关论文
共 24 条
[1]  
Chung K. L., 1968, A course in probability theory
[2]  
GUO L, 1994, IEEE DECIS CONTR P, P3894, DOI 10.1109/CDC.1994.411776
[3]   ESTIMATION OF PREDICTION ERROR VARIANCE [J].
HANNAN, EJ ;
NICHOLLS, DF .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1977, 72 (360) :834-840
[4]   ASYMPTOTIC PROPERTIES OF NON-LINEAR LEAST SQUARES ESTIMATORS [J].
JENNRICH, RI .
ANNALS OF MATHEMATICAL STATISTICS, 1969, 40 (02) :633-&
[5]  
KOSUT RL, 1992, IEEE T AUTOMATIC CON, V37
[6]  
KOUST RL, 1992, IEEE T AUTOMAT CONTR, V37, P929
[8]   ASYMPTOTIC PROPERTIES OF BLACK-BOX IDENTIFICATION OF TRANSFER FUNCTIONS. [J].
Ljung, Lennart ;
Yuan, Zhen-Dong .
IEEE Transactions on Automatic Control, 1985, AC-30 (06) :514-530
[9]  
Ljung L., 1979, Stochastics, V3, P29
[10]  
Ljung L., 1999, SYSTEM IDENTIFICATIO