Improved robust stability criteria for neural networks with fast time-varying delays

被引：1

作者：

Mahmoud, M. S. ^{[1
]}

Al-Rayyan, A. Y. ^{[1
]}

Xia, Y. ^{[2
]}

机构：

[1] King Fahd Univ Petr & Minerals, Dept Syst Engn, Dhahran 31261, Saudi Arabia

[2] Beijing Inst Technol, Dept Automat Control, Beijing 100081, Peoples R China

来源：

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART I-JOURNAL OF SYSTEMS AND CONTROL ENGINEERING | 2010年 / 224卷 / I5期

关键词：

delay-dependent stability; neural networks; time-varying delay; linear matrix inequality; GLOBAL ASYMPTOTIC STABILITY; DEPENDENT EXPONENTIAL STABILITY; STATE ESTIMATION; SYSTEMS;

D O I：

10.1243/09596518JSCE943

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

An improved robust global stability criterion is developed for uncertain neural networks with fast time-varying delays. The networks have norm-bounded parametric uncertainties. The relationship between the time-varying delay and associated extreme bounds (lower and upper) is appropriately exploited when dealing with the Lyapunov functional derivative. The developed stability criterion is delay dependent and is characterized by linear-matrix-inequality-based conditions. Numerical examples are presented to illustrate the benefits and lower conservativeness of the developed method.

引用

页码：521 / 528

页数：8