Complete Delay-Decomposing Approach to Asymptotic Stability for Neural Networks With Time-Varying Delays.
- Resource Type
- Article
- Authors
- Zeng, Hong-Bing; He, Yong; Wu, Min; Zhang, Chang-Fan
- Source
- IEEE Transactions on Neural Networks. 05/01/2011, Vol. 22 Issue 5, p806-812. 7p.
- Subject
- *ARTIFICIAL neural networks
*TIME delay systems
*MATHEMATICAL decomposition
*STABILITY (Mechanics)
*LYAPUNOV functions
*NUMERICAL analysis
*SYMMETRIC matrices
- Language
- ISSN
- 1045-9227
This paper is concerned with the problem of stability of neural networks with time-varying delays. A novel Lyapunov–Krasovskii functional decomposing the delays in all integral terms is proposed. By exploiting all possible information and considering independent upper bounds of the delay derivative in various delay intervals, some new generalized delay-dependent stability criteria are established, which are different from the existing ones and improve upon previous results. Numerical examples are finally given to demonstrate the effectiveness and the merits of the proposed method. [ABSTRACT FROM AUTHOR]