作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 人工智能及识别技术 • 上一篇    下一篇

基于忆阻器时滞神经网络的全局稳定性新判据

段飞腾,崔宝同   

  1. (江南大学物联网工程学院,江苏 无锡 214122)
  • 收稿日期:2014-11-04 出版日期:2015-07-15 发布日期:2015-07-15
  • 作者简介:段飞腾(1990-),男,硕士研究生,主研方向:神经网络;崔宝同,教授、博士生导师。

New Criterion for Global Stability Based on Memristor Time Delay Neural Network

DUAN Feiteng,CUI Baotong   

  1. (School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,China)
  • Received:2014-11-04 Online:2015-07-15 Published:2015-07-15

摘要: 针对具有时滞的忆阻器神经网络,研究全局一致渐近稳定性问题,提出基于M-矩阵的充分性判据。通过构造Lyapunov泛函,采用同胚映射原理和微分包含的研究方法,推导证明一类时滞忆阻器神经网络的平衡点存在性和唯一性,并说明系统的平衡点是全局渐近稳定,所得判据扩展了基于M-矩阵的结果,对于不同的时滞和激活函数具有一定的鲁棒性,并且判据根据系统的本身物理参数即可以验证。数值分析与仿真结果验证了新判据的有效性。

关键词: 忆阻器, 稳定性, 神经网络, 时滞, M-矩阵

Abstract: This paper researches global exponential stability of memristor-based recurrent neural networks with time delays.By employing homeomorphism mapping,Lyapunov functionals and differential theory,the existence and uniqueness of the equilibrium point of memristor-based neural networks are proved and the equilibrium is global asymptotic stability,some M-matrix based conditions are obtained.In addition,the conditions improve some previous criteria based on M-matrix and have robustness for different time delays and activation function.Besides,the conditions are easy to be verified with the physical parameters of system itself.Numerical analysis and simulation result shows the effectiveness of the new criterion.

Key words: memristor, stability, neural networ, time delay, M-martix

中图分类号: