具有参数摄动时滞Hopfield神经网络的鲁棒稳定性和吸引域估计  

A Robust Stability and Estimation of Attraction Domain for Hopfield Neural Networks with Parameter Perturbation and Tim Delay

作　　者：李淑侠[1] 王占山[2] 

机构地区：[1]朝阳师范高等专科学校,辽宁朝阳122000 [2]沈阳理工大学

出　　处：《沈阳理工大学学报》2007年第1期21-24,共4页Journal of Shenyang Ligong University

摘　　要：针对具有参数摄动的时滞Hopfield神经网络,基于矩阵范数的概念建立了时滞依赖鲁棒稳定判据.该鲁棒稳定判据将参数摄动大小和时滞上界大小统一在一个表达式中,进而可得到时滞和参数摄动大小对网络稳定性影响的定量关系.此外,给出了局部稳定平衡点吸引域大小的估计,并通过数值仿真验证了本文结果的有效性.The delay-dependant robust stability for delayed Hoptield neural networks with asymmetric interconnection structure due to parameter perturbations is analyzed via matrix norm approach. The obtained result combines the time delay and the parameter perturbation into one expression, which shows the quantitative effects of time delay and parameter perturbation on the robust stability of Hopfield neural networks. Further, a procedure of estimating the domain of attraction of equilibrium point of Hopfield neural networks is established. A numerical example is used to prove the effectiveness of the proposed approaches.

关 键 词：HOPFIELD神经网络 参数摄动 鲁棒稳定 吸引域 时滞 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]