摘要
研究了Cohen-Grossberg神经网络模型的指数稳定性.运用非线性测度方法证明了神经网络平衡点的存在性和惟一性,接着通过构造一个新颖的Lyapunov泛函,得到了神经网络指数稳定的全新充分条件,并给出了解的指数衰减的精确估计.与已有文献相比,文中给出的条件更为宽松且易于验证.
The exponential stability ot Cohen-Grossberg neural net work model is considered. The nonlinear measure approach is employed to analyze the existence and uniqueness of an equilibrium. A novel Lyapunov functional is constructed to derive the stability of neural networks. New sufficient conditions for the existence of a unique equilibrium and the exponential stability of the neural networks are presented. Moreover, the exponential decay estimate of the solution is precisely characterized. The proposed criteria are milder and more flexible to verify than the existing results.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2006年第2期215-218,共4页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(10371097)
关键词
神经网络
指数稳定性
指数衰减估计
非线性测度
neural network
exponential stability
exponential decay estimate
nonlinear measure