离散递归神经网络平衡点控制及稳定性分析  

Control and Stability Analysis of Equilibrium of Discrete-Time Recurrent Neural Networks

作　　者：柯云泉[1] 

机构地区：[1]绍兴文理学院数学系,浙江绍兴312000

出　　处：《绍兴文理学院学报》2007年第8期18-26,共9页Journal of Shaoxing University

基　　金：浙江省自然科学基金资助项目(Y604003)

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.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.

关 键 词：离散递归神经网络 状态空间 平衡点 稳定性 

分 类 号：O175.13[理学—数学]