具有时滞的脉冲随机神经网络的有限时间稳定性  

Finite-time stability of the impulsive stochastic neural networks with delay

作　　者：鲁成甜 喻圣 程培[1] LU Cheng-tian;YU Sheng;CHENG Pei(School of Mathematical Sciences,Anhui University,Hefei Anhui 230601,China)

机构地区：[1]安徽大学数学科学学院

出　　处：《控制理论与应用》2020年第1期187-192,共6页Control Theory & Applications

基　　金：国家自然科学基金项目(11771001);安徽省高等学校自然科学研究项目(KJ2018A0027)资助~~

摘　　要：本文研究具有时滞的脉冲随机神经网络的有限时间稳定性问题.利用Lyapunov泛函技术,线性矩阵不等式(LMIs)工具和平均脉冲区间条件,对反镇定型、中立型和镇定型3种类型的脉冲系统分别给出了基于矩阵不等式的有限时间均方稳定的充分条件,最后通过一个数值例子验证了理论结果的有效性.This paper focuses on the problem of finite-time stability of the impulsive stochastic neural networks with delay.Three types of impulses are considered:the impulses are input disturbances,the impulses are neutral type and the impulses are stabilizing.For each type of impulses,by using Lyapunov functional and linear matrix inequalities(LMIs)techniques combined with the concept of the average impulsive interval,the sufficient conditions for the mean square finitetime stability are established in terms of matrix inequalities.Finally,a numerical example is given to verify the effectiveness of the theoretical results.

关 键 词：神经网络 脉冲 线性矩阵不等式 LYAPUNOV泛函 有限时间稳定 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O231[自动化与计算机技术—控制科学与工程]