基于非连续控制协议的切换耦合神经网络有限时间二分同步  

Finite-time bipartite synchronization of switched coupled neural networks based on discontinuous control protocols

作　　者：梁军丽 胡元发 刘小洋 严浩远 Liang Junli;Hu Yuanfa;Liu Xiaoyang;Yan Haoyuan(School of Computer Science and Technology,Jiangsu Normal University,Xuzhou 221116,China)

机构地区：[1]江苏师范大学计算机科学与技术学院,江苏徐州221116

出　　处：《南京理工大学学报》2021年第3期306-313,共8页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(61773185,61877030);2019年江苏省高校“青蓝工程”。

摘　　要：该文考虑了符号图下具有非连续激活函数的切换耦合神经网络有限时间二分同步问题。基于Lyapunov稳定性理论,针对结构平衡图和结构非平衡图,分别设计了相应的非连续控制协议,获得了耦合网络实现有限时间二分同步的判别准则。借助微分包含与集值李导数理论,保证了耦合非连续神经网络在非连续控制协议下的有限时间同步,获得了收敛时间的上界,提高了同步网络的鲁棒性与抗干扰性能。最后通过2个仿真实例,验证了理论结果的有效性和可行性。The problem of finite-time bipartite synchronization of switched coupled neural networks with discontinuous activation function is considered in signed graphs.Based on the Lyapunov stability theory,two discontinuous control protocols are constructed for the structurally balanced and structurally unbalanced graphs,and some sufficient criteria are achieved for the coupled networks to realize finite-time bipartite synchronization.The finite-time synchronization of coupled networks under discontinuous control protocols is insured with the help of differential inclusion and set-valued Lie derivative,and the corresponding settling time is estimated.Then,the robustness and disturbance rejection of synchronized network is ensured as well.Finally,the validity and feasibility of the theoretical results are verified by two simulation examples.

关 键 词：非连续控制协议 耦合神经网络 有限时间 二分同步 符号图 微分包含 集值李导数 

分 类 号：O193[理学—数学]