含有分布时滞和脉冲的杂交BAM神经网络的全局指数稳定性  

An analysis on global exponential stability of hybrid BAM neural networks with distributed delays and impulse

作　　者：王继禹[1] 贾秀玲[1] 

机构地区：[1]河南理工大学万方科技学院公共基础部,河南郑州451400

出　　处：《南阳师范学院学报》2011年第6期1-7,共7页Journal of Nanyang Normal University

基　　金：国家自然科学基金资助项目(30770500)

摘　　要：通过构造李雅普诺夫函数,利用M-矩阵理论以及Young不等式技巧,给出了一类含有分布时滞和脉冲双向联想记忆神经网络的平衡点的全局指数稳定性的充分条件,这些条件去掉了对激活函数的有界性、单调性和可微性的要求,且在某些情况下更易验证.By utilizing the Lyapunov functional method,applying the M-matrix theory,Young inequality technique and some analysis techniques,some sufficient conditions is obtained for the global exponential stability of equilibrium point for hybrid bidirectional associative memory（BAM） neural networks and subjected to impulsive state displacements at fixed instants of time.The results remove the assumptions that the boundedness,monotonicity or differentiability of the activation functions.It is shown that in some cases,the stability criteria can be easily checked.

关 键 词：全局指数稳定性 双向联想记忆神经网络 M-矩阵 

分 类 号：O175.12[理学—数学]