电子学报

• 学术论文 • 上一篇    下一篇

具比例时滞杂交双向联想记忆神经网络的全局指数稳定性

周立群   

  1. 天津师范大学数学科学学院, 天津 300387
  • 收稿日期:2012-07-20 修回日期:2013-08-06 出版日期:2014-01-25
    • 作者简介:
    • 周立群 女,1972年9出生,黑龙江省齐齐哈尔市人,副教授.2004年与2007年在哈尔滨工业大学获得理学硕士和工学博士学位.主要从事神经网络的理论及应用研究. E-mail:zhouliqun20000@163.com
    • 基金资助:
    • 国家自然科学基金 (No.61374009); 天津市高等学校科技发展基金 (No.20100813); 天津师范大学博士基金 (No.52LX34)

Global Exponential Stability in Hybrid Bidirectional Associative Memory Neural Networks with Proportional Delays

ZHOU Li-qun   

  1. School of Mathematics Science, Tianjin Normal University, Tianjin 300387, China
  • Received:2012-07-20 Revised:2013-08-06 Online:2014-01-25 Published:2014-01-25
    • Supported by:
    • National Natural Science Foundation of China (No.61374009); Science and Technology Development Fund of Colleges and Universities in Tianjin Municipality (No.20100813); Doctoral Fundation of Tianjin Normal University (No.52LX34)

摘要: 对一类具比例时滞杂交双向联想记忆神经网络进行研究,利用Brouwer不动点定理证明该网络的平衡点的存在唯一性.利用变换将具比例时滞杂交双向联想记忆神经网络变换成等价的具不等常时滞与变系数杂交双向联想记忆神经网络.利用不等式技巧建立一拟Halanay型不等式系统,进而得到了确保该系统全局指数稳定的时滞独立的充分条件.并给出两个算例验证所得结论的正确性.

关键词: 联想记忆, 神经网络, 比例时滞, Brouwer不动点定理, 全局指数稳定性

Abstract: In this paper,a class of hybrid bidirectional associative memory neural networks with proportional delays is studied.In virtue of Brouwer fixed point theorem,the existence and uniqueness of equilibrium of the system are proved.The transformation transforms the neural networks with proportional delays into the neural networks with unequal constant delays and variable coefficients.By applying inequality techniques,a system of the quasi Halanay-type inequalities is established,and a delay-independent sufficient condition is derived for ensuring the global exponential stability of equilibrium of the system.And two examples are given to illustrate the effectiveness of obtained results.

Key words: bidirectional associative memory, neural networks, proportional delays, Brouwer fixed point theorem, global exponential stability

中图分类号: