文献类型：期刊文献

中文题名：离散递归神经网络平衡点控制及稳定性分析

英文题名：Control and Stability Analysis of Equilibrium of Discrete-Time Recurrent Neural Networks

作者：柯云泉[1]

机构：[1]绍兴文理学院数学系

年份：2007

卷号：27

期号：8

起止页码：18

中文期刊名：绍兴文理学院学报

收录：国家哲学社会科学学术期刊数据库

基金：浙江省自然科学基金资助项目(Y604003)

语种：中文

中文关键词：离散递归神经网络;状态空间;平衡点;稳定性

外文关键词：Discrete - time recurrent neural networks ; state space ; equilibrium ; stability

中文摘要：通过将离散递归神经网络的状态空间分割为若干个子区域,控制其输出反馈权值、输入项及阈值等,分析在每一个子区域上平衡点的存在、唯一性问题,给出了其判定条件.并进一步讨论了平衡点渐近稳定、指数稳定性.通过举例说明给定的结果简明又容易验证.

外文摘要：By dividing state space into some sub - domains and controlling the output synaptic weights, input terms and the threshold, we analyzed the existence and uniqueness of the equilibrium of discrete - time recurrent neural networks, giving some sufficient conditions for the existence of equilibrium in some sub - domains. Then the asymptotic stability and exponential stability of the equilibrium were investigated, giving some examples to illustrate that the resuits are concise and easy to operate.