Passivity Analysis of Neural Networks With Time Delay

Author

Li, Chunguang ; Liao, Xiaofeng

Author_Institution

Centre for Nonlinear & Complex Syst., Univ. of Electron. Sci. & Technol. of China, Chengdu, China

Volume

52

Issue

8

fYear

2005

Firstpage

471

Lastpage

475

Abstract

The passivity conditions for delayed neural networks (DNNs) are considered in this paper. We firstly derive the passivity condition for DNNs without uncertainties, and then extend the result to the case with time-varying parametric uncertainties. The proposed approach is based on a Lyapunov–Krasovskii functional construction. The passivity conditions are presented in terms of linear matrix inequalities, which can be easily solved by using the effective interior-point algorithm. Numerical examples are also given to demonstrate the effectiveness of the theoretical results.

Keywords

Lyapunov matrix equations; analogue circuits; circuit stability; delay circuits; linear matrix inequalities; network analysis; neural nets; DNN; LMI; Lyapunov-Krasovskii functional construction; delayed neural networks; interior-point algorithm; linear matrix inequality; passivity analysis; time delay; time-varying parametric uncertainty; Artificial neural networks; Circuit stability; Delay effects; Linear matrix inequalities; Multi-layer neural network; Neural networks; Process control; Signal processing algorithms; Stability analysis; Uncertainty; Delay; linear matrix inequality (LMI); neural networks; passivity; uncertainty;

fLanguage

English

Journal_Title

Circuits and Systems II: Express Briefs, IEEE Transactions on

Publisher

ieee

ISSN

1549-7747

Type

jour

DOI

10.1109/TCSII.2005.849023

Filename

1495752