基于非平滑分析的时变时滞神经网络系统的全局渐近稳定性研究

Research on Global Asymptotic Stability of the Delayed Neural Network with Nonsmooth Analysis

基于Lyapunov稳定性理论,利用非平滑分析的方法去掉了激励函数可微和有界的条件,通过构造一种新颖的Lyapunov泛函推广改善了判定时滞神经网络系统的全局稳定性的条件.数值仿真表明了结果的有效性.

Based on Lyapunov stability theory, this paper is concerned with the global asymptotic stability problem of the delayed neural network by constructing a novel Lyapunov functional. And this paper generalizes the global asymptotic stability conditions of neural network system by removing the differentiable and bounded conditions of activation function with the nonsmooth Analysis. The simulation example is given to illustrate the effectiveness of the results.