东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2009, Vol. 30 ›› Issue (5): 613-616.DOI: -

• 论著 • 上一篇    下一篇

一类时变时滞静态神经网络的指数稳定性

张锐;王占山;井元伟;   

  1. 东北大学信息科学与工程学院;
  • 收稿日期:2013-06-22 修回日期:2013-06-22 出版日期:2009-05-15 发布日期:2013-06-22
  • 通讯作者: Zhang, R.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(60774093);;

Exponential stability of a class of static neural networks with time varying delay

Zhang, Rui (1); Wang, Zhan-Shan (1); Jing, Yuan-Wei (1)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China
  • Received:2013-06-22 Revised:2013-06-22 Online:2009-05-15 Published:2013-06-22
  • Contact: Zhang, R.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

660

摘要： 基于线性矩阵不等式方法建立了时变时滞静态神经网络的指数稳定判据.考虑到时滞变化率对稳定性能的影响,分别建立了仅依赖时滞上界的稳定判据和完全依赖时滞信息的稳定判据.所得到的稳定判据能够适应慢变时滞和快变时滞两种情况,具有适用范围宽、保守性小和易于验证等特点,并通过几个注释说明与现有的文献结果进行了比较.仿真示例验证了所得结果的有效性.

关键词: 递归神经网络, 局部场神经网络, 静态神经网络, Lyapunov泛函, 全局指数稳定, 线性矩阵不等式

Abstract: Based on the linear matrix inequalities and considering the different effects of change rate of time varying delay on stability, two criteria for exponential stability are set up, i.e. the criterion that just depends on the upper bound of the delay and the criterion that depends wholly on the delay information. The two stability criteria obtained can be adaptable to either quickly changed or slowly changed time varying delay, thus providing wider applications, less conservativeness and verification easy to do, and with some remarks given and in comparison to the results in earlier works, simulations were done to exemplify the effectiveness of the proposed approaches.

中图分类号: