摘要
在激励函数满足全局Lipshitz条件和时变时滞函数有界的情形下,研究了一类具有时变时滞和无穷分布时滞的BAM神经网络模型.首先利用压缩映像原理给出了系统平衡点的存在唯一性条件;然后利用初等积分法并结合矩阵论知识,不需构造Lyapunov-Krasovskii泛函,以矩阵谱半径的形式给出了模型平衡点全局指数稳定的充分条件,还以矩阵范数的形式给出了更为严格但极易验证的充分条件.结论去掉了对激励函数有界性和时变时滞函数连续可微性,甚至导数具有不超过1的上界等严格要求,推广改进了相关文献的结果.最后举例说明了本文方法的有效性.
Under the circumstance that the activation functions meet global Lipschitz conditions and the time-varying delay is bounded,the BAM neural network with time-varying and continuously infinite distributed time delays is studied in this paper.Firstly by means of contraction reflection principle,a sufficient condition for the existence and uniqueness of the equilibrium point is derived.Then by employing the matrix theory and elementary method of integration,without constructing a Lyapunov-Krasovskii functional,sufficient conditions for guaranteeing the globally exponential stability of the equilibrium point are obtained in the form of linear matrix inequality（LMI）.Another sufficient condition which is more strict but much easier to be checked is also derived in the form of matrix norm.The boundedness of the activation function,the continuous differentiability of the time-varying delay and even the upper bound of its derivative which is no larger than 1,are removed.The results improved and extended the previous ones.An example is also given to illustrate the effectiveness of this proposed method in this paper.
出处
《哈尔滨理工大学学报》
CAS
2012年第6期6-13,共8页
Journal of Harbin University of Science and Technology
基金
海军航空工程学院专业技术拔尖人才基金(名师工程)
关键词
BAM神经网络
压缩映像原理
时变时滞
连续分布时滞
全局指数稳定
BAM neural networks
contraction reflection principle
time-varying time-delay
continuously distributed time-delay
globally exponentially stable
