摘要
虚拟化网络分析有利于解决复杂系统的控制问题,但是现有虚拟化网络研究主要针对非奇异网络,无法有效应用于包含奇异节点的网络处理。为解决上述问题,基于节点间的多重关联关系构造了奇异网络模型,分析领导节点以及网络偏差模型。基于模型分析设计了牵制控制算法,利用节点状态以及彼此存在的关联关系,通过牵制节点的改变来实现对全部节点的整体控制。根据偏差系统推导出反馈控制器,利用Laplacian矩阵确定牵制节点,并利用克罗内克积与线性微分计算得到合理的反馈增益系数,从而确保反馈牵制控制效果。通过算例仿真,方法能够针对节点的奇异性准确构造虚拟化网络模型,并根据网络关联性与密度的具体情况对牵制控制强度进行调整,实现系统的快速有效同步控制,显著降低了系统控制的偏差,提高了系统控制的鲁棒性和平稳性。
Virtual network analysis is helpful to solve the control problem of a complex system.However,the existing virtualization network research mainly focuses on nonsingular networks,which can not be effectively applied to the network processing with singular nodes.Therefore,this paper constructs a singular network model based on the multiple association relationship between nodes,analyzes the leader node and the network deviation model.The pinning control algorithm was designed based on the model analysis.Using the state of nodes and the relationship between them,the overall control of all nodes was achieved by controlling the changes of nodes.The feedback controller was derived from the deviation system,and the pinning nodes were determined by Laplacian matrix.The reasonable feedback gain coefficient was obtained by using the Kronecker product and linear differential calculation,so as to ensure the feedback control effect.Through the simulation of an example,the method can accurately construct the virtual network model according to the singularity of nodes.According to the specific situation of network relevance and density,the intensity of containment control is adjusted,the fast and effective synchronization control of the system is realized,the deviation of the system control is significantly reduced,and the robustness and stability of the system control are improved.
作者
孟凡姿
侯云海
MENG Fan-zi;HOU Yun-hai(Jilin University of Architecture and Technology,Changchun Jilin 130011,China;Changchun Unversty of Technology,Changchunjilin 130011,China)
出处
《计算机仿真》
北大核心
2021年第10期324-327,423,共5页
Computer Simulation
基金
吉林省教育厅“十二五”项目(吉教科合字[2015]第98号)。
关键词
虚拟化网络
复杂系统
奇异节点
反馈牵制控制
增益系数
Virtual network
Complex system
Singular node
Feedback pinning control
Gain coefficient
