Novel synchronization criteria for a class of neural networks with multiple delays

详细信息    查看官网全文

• 英文论文题名：Novel synchronization criteria for a class of neural networks with multiple delays
• 论文作者：Ge Juhong ; Xu Jian
• 英文论文作者：Ge Juhong ; Xu Jian ; Department of Mathematics and Information Science ; He'nan University of Economics and Law ; School of Aerospace Engineering and Applied Mechanics ; Tongji University
• 年：2016
• 作者机构：Department of Mathematics and Information Science, He'nan University of Economics and Law;School of Aerospace Engineering and Applied Mechanics, Tongji University;
• 英文论文关键词：multiple delays ; synchronous periodic solution ; perturbation-incremental scheme
• 会议召开时间：2016-05-06
• 会议录名称：第十届动力学与控制学术会议摘要集
• 语种：英文
• 分类号：TP183
• 学会代码：AGLU
• 会议名称：第十届动力学与控制学术会议
• 会议地点：中国四川成都
• 主办单位：中国力学学会动力学与控制专业委员会
• 学会名称：中国力学学会
• 页数：2
• 文件大小：744k
• 原文格式：D
• 会议级别：全国

摘要

In this paper, a general artificial neural network model is studied for four neurons with bidirectional time delayed connections between the neurons. We try to find out some critical condition under which synchronous periodic oscillation occur when the parameters of the networkmodel pass through some critical values. We present a detailed discussion about synchronous periodic solution of the network by employing the perturbation-incremental scheme. The synchronous periodic solution arising from Hopf bifurcation due to time delay is analytically given, at the same time, the necessary and sufficient condition for the synchronous periodic solution are also obtained. The results obtained in this paper give an explicit view of the dynamics of the delayed neural network. It can be seen that theoretical analysis is in good agreement with the numerical simulation, which shows that the obtained result is correct. To our knowledge, obtaining analytically synchronous periodic solution by employing the perturbation-incremental scheme is new in the literature.
In this paper, a general artificial neural network model is studied for four neurons with bidirectional time delayed connections between the neurons. We try to find out some critical condition under which synchronous periodic oscillation occur when the parameters of the networkmodel pass through some critical values. We present a detailed discussion about synchronous periodic solution of the network by employing the perturbation-incremental scheme. The synchronous periodic solution arising from Hopf bifurcation due to time delay is analytically given, at the same time, the necessary and sufficient condition for the synchronous periodic solution are also obtained. The results obtained in this paper give an explicit view of the dynamics of the delayed neural network. It can be seen that theoretical analysis is in good agreement with the numerical simulation, which shows that the obtained result is correct. To our knowledge, obtaining analytically synchronous periodic solution by employing the perturbation-incremental scheme is new in the literature.

引文