摘要
研究了基于凸包型脉冲的时标上多重循环神经网络系统(multiple recurrent neural networks,MRNNs)的全局同步问题.利用序列连通、时标理论和凸包型脉冲控制策略得到了混合时域上保证全局同步的充分条件.主要结果不仅有一般性,也推广了已有文献的相关结论,其中时标理论确保了得到的充分条件不仅可应用于连续时域和离散时域,还可应用在混合时域上.最后用一个例子验证了结论的正确性.
We study a global synchronization problem of multiple recurrent neural networks based on convex-type impulsive control on time scales.Sufficient conditions are derived to achieve the global synchronization on hybrid time domains by using the concept of sequential connectivity and the theory of time scales with convex-type impulsive control scheme.The main result not only secures the universality,but also generalizes relevant conclusions of existing works.Furthermore,the application theory of time scales ensures that these derived conditions can be applied to the continuous-time domain and the discrete-time domain as well as to the hybrid time domain.Finally,an example demonstrates and confirms our claim.
作者
蔡庆瑞
黄振坤
CAI Qingrui;HUANG Zhenkun(School of Sciences,Jimei University,Xiamen 361021,China)
出处
《厦门大学学报（自然科学版）》
CAS
CSCD
北大核心
2022年第2期186-193,共8页
Journal of Xiamen University：Natural Science
基金
国家自然科学基金(61573005)
福建省自然科学基金(2019J01330)。
关键词
循环神经网络
同步
脉冲控制
序列连通
时标
multiple recurrent neural network
synchronization
impulsive control
sequential connectivity
time scale
