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.
出处
《沈阳理工大学学报》
CAS
2007年第1期21-24,共4页
Journal of Shenyang Ligong University
关键词
HOPFIELD神经网络
参数摄动
鲁棒稳定
吸引域
时滞
Hopfield neural networks
parameter perturbations
robust stability
domain of attraction
time delay
