Mixed parametric/non-parametric identification of systems with discontinuous nonlinearities

被引：8
|
作者：
Vincent, Tyrone L. [1 ]
Novara, Carlo [2 ]
机构：
[1] Colorado Sch Mines, Dept Elect Engn & Comp Sci, Golden, CO 80401 USA
[2] Politecn Torino, Dipartimento Automat & Informat, Turin, Italy
来源：
AUTOMATICA | 2013年 / 49卷 / 12期
关键词：
System identification; Nonlinear systems; Discontinuities; Regularization;
D O I：
10.1016/j.automatica.2013.09.022
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
The subject of this paper is identification of discrete time nonlinear dynamical systems when the system dynamics are defined by a discontinuous nonlinear function with the location of the discontinuity unknown. By representing the nonlinear function using both a parametric term to capture the continuous part and a non-parametric term to capture the discontinuous part, we present an identification algorithm along with conditions for recovery of the true nonlinearity. (C) 2013 Elsevier Ltd. All rights reserved.
引用
收藏
页码：3661 / 3669
页数：9
相关论文
共 50 条
12345