摘要
针对具有时变时滞的中立型Hopfield神经网络,在对时变时滞的合理假设下,文中通过构造适当的Lyapunov泛函,提出了一种新的方法判定时变时滞中立型Hopfield神经网络系统的全局一致渐近稳定性。所得判据是建立在一个代数Riccati方程及其对应离散Lyapunov方程存在对称正定解之上的,并得出了易于计算机仿真实现的线性矩阵不等式的判据形式,仿真验证了结果的有效性。
A method is proposed for uniformly asymptotic stability analysis of neutral neural Hopfield networks with time-varying delay.The delay is assumed unknown,time varying but the varying speed bounded by a constant less than one.Sufficient conditions on delay-independtent uniformly asymptotic stability are given in terms of the existence of symmetric and positive definite solutions of a constant Riccati algebraic matrix equation coupled with a discrete Lyapunov equation,the corresponding LMI formulation for the stability creterion is also derived.The simulation results show the effectiveness of the given stability creterion.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2010年第6期136-139,共4页
Journal of Wuhan University of Technology
基金
国家自然科学基金(60974136)
