基于LMI的时滞细胞神经网络的全局渐近稳定性分析  

Generalized LMI-Based Approach to the Global Asymptotic Stability of Cellular Neural Networks With Delay

作　　者：刘德友[1] 张建华[1] 关新平[1] 肖晓丹[1] 

机构地区：[1]燕山大学理学院,河北秦皇岛066004

出　　处：《应用数学和力学》2008年第6期735-740,共6页Applied Mathematics and Mechanics

基　　金：国家自然科学基金资助项目(60604004);河北省自然科学基金资助项目(F2007000637);国家杰出青年自然科学基金资助项目(60525303)

摘　　要：研究了一类具有时滞的细胞神经网络的稳定性问题,利用Liapunov-Krasovskii泛函的方法,给出了时滞相关的稳定性判据.稳定性判据是以线性矩阵不等式(LMI)的形式给出,可以很容易得出时滞的上界.在得到时滞相关的稳定性判据的同时也可以得到时滞无关的稳定性判据,包含了已有文章中的很多结果.最后,数值算例说明了结果的优越性.The global asympototic stability problem of cellular neural networks with delay is investigated. A new stability condition was presented based on Liapunov-Krasovskii method, which is dependent on the size of delay. The result is given in the form of LMI（linear matric inequality） ,and the admitted tipper bound of the delay can be obtained easily. The time delay dependent and independent resuits can be obtained, which include some results in the former literature. Finally, a numerical exam- ple was given to illustrate the effectiveness of the main resttlts.

关 键 词：时滞细胞神经网络 线性矩阵不等式 全局稳定 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]