Exponential stability of uncertain stochastic neural networks with mixed time-delays
Exponential stability of uncertain stochastic neural networks with mixed time-delays
复制标题
DOI：
10.1016/j.chaos.2005.10.061
复制
发表时间：
2007-04-01
影响因子：
7.8
通讯作者：
Liu, Xiaohui
中科院分区：
文献类型：
--
作者：
Wang, Zidong;Lauria, Stanislao;Liu, Xiaohui
研究了一类具有混合时滞和参数不确定性的随机神经网络的全局指数稳定性分析问题。混合时滞包括离散和分布时滞，参数不确定性是范数有界的，神经网络受到布朗运动描述的随机扰动。稳定性分析问题的目的是推导出易于测试的标准，在该标准下，对于所有可允许的参数不确定性，延迟随机神经网络都是全局、鲁棒、均方指数稳定的。借助于Lyapunov-Krasovskii稳定性理论和随机分析工具，利用一种有效的线性矩阵不等式（LMI）方法，建立了系统稳定性的充分条件.建议的标准可以很容易地检查，使用最近开发的数值包，其中没有调整的参数是必需的。一个例子来证明所提出的标准的实用性。(c)2005爱思唯尔有限公司保留所有权利。
This paper is concerned with the global exponential stability analysis problem for a class of stochastic neural networks with mixed time-delays and parameter uncertainties. The mixed delays comprise discrete and distributed time-delays, the parameter uncertainties are norm-bounded, and the neural networks are subjected to stochastic disturbances described in terms of a Brownian motion. The purpose of the stability analysis problem is to derive easy-to-test criteria under which the delayed stochastic neural network is globally, robustly, exponentially stable in the mean square for all admissible parameter uncertainties. By resorting to the Lyapunov-Krasovskii stability theory and the stochastic analysis tools, sufficient stability conditions are established by using an efficient linear matrix inequality (LMI) approach. The proposed criteria can be checked readily by using recently developed numerical packages, where no tuning of parameters is required. An example is provided to demonstrate the usefulness of the proposed criteria. (c) 2005 Elsevier Ltd. All rights reserved.