变时滞Hopfield神经网络的全局指数稳定性  

EXPONENTIAL STABILITY OF HOPFIELD NEURAL NETWORKS WITH TIME-VARYING DELAYS

作　　者：翁良燕[1] 周立群[1] 

机构地区：[1]天津师范大学数学科学学院,天津300387

出　　处：《数学杂志》2013年第5期909-915,共7页Journal of Mathematics

基　　金：国家自然科学基金项目(60974144);天津师范大学博士基金项目(52LX34);天津高等学校科技发展基金项目(20100813)

摘　　要：本文研究了一类具有变时滞Hopfield神经网络的稳定性问题.利用时滞微分不等式方法,获得了几个关于该网络的全局指数稳定性与时滞无关的充分条件,并且给出了此类网络的收敛指数的估计,推广了已知文献的结果.最后给出数值例子证明结论的有效性.Global exponential stability of Hopfield neural networks with time-varying delays is investigated. Base on the delay differential inequality approach, some delay-independent sufficient conditions are obtained for the exponential stability of neural networks, moreover, given the exponential convergence of equilibrium states, these results are better than ones in the references, two examples are given to illustrate the effectiveness of the obtained results.

关 键 词：HOPFIELD神经网络 变时滞 全局指数稳定性 时滞微分不等式 

分 类 号：O175.13[理学—数学]