APPROXIMATE GRADIENTS, CONVERGENCE AND POSITIVE REALNESS IN RECURSIVE-IDENTIFICATION OF A CLASS OF NONLINEAR-SYSTEMS

被引:21
作者
WIGREN, T
机构
[1] Systems and Control Group, Department of Technology, Uppsala University, Uppsala
关键词
ADAPTIVE FILTERING; CONVERGENCE ANALYSIS; NONLINEAR SYSTEMS; QUANTIZATION;
D O I
10.1002/acs.4480090403
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Recursive identification algorithms based on the Wiener model are presented in this paper. They estimate the parameters of a SISO linear dynamic block in cascade with a known static output non-linearity. Inversion of the non-linear function is avoided and approximations of gradients are utilized. This allows an exact treatment of output measurement saturation and of situations where output measurements are obtained from sensors with relay-type characteristics, such as EGO sensors in emission control systems for cars. Exact compensation for coarse quantization of output measurements can also be obtained by the algorithms. Stochastic averaging techniques using associated differential equations prove that local and global convergence of the schemes are tied to positive realness and sector conditions on the non-linearity. Conditions for local convergence to the correct parameters are established for the case where the output non-linearity is an arbitrary quantizer.
引用
收藏
页码:325 / 354
页数:30
相关论文
共 41 条
[1]  
Astrom K., 1997, COMPUTER CONTROLLED, V3
[2]  
Astrom K.J., 1995, ADAPTIVE CONTROL
[3]   ZEROS OF SAMPLED SYSTEMS [J].
ASTROM, KJ ;
HAGANDER, P ;
STERNBY, J .
AUTOMATICA, 1984, 20 (01) :31-38
[4]  
Benveniste A., 1987, International Journal of Adaptive Control and Signal Processing, V1, P3, DOI 10.1002/acs.4480010103
[5]  
Billings S., 1980, P I ELECT ENG D, V127, P272, DOI DOI 10.1049/IP-D:19800047
[6]   IDENTIFICATION OF SYSTEMS CONTAINING LINEAR DYNAMIC AND STATIC NON-LINEAR ELEMENTS [J].
BILLINGS, SA ;
FAKHOURI, SY .
AUTOMATICA, 1982, 18 (01) :15-26
[7]  
BILLINGS SA, 1978, ELECTRON LETT, V13, P502
[9]  
DING Z, 1990, 29TH P IEEE C DEC CO, P225
[10]  
DOEBLIN EQ, 1983, MEASUREMENT SYSTEMS