有向复杂网络的目标可控性分析

Analysis of Target Controllability of Directed Complex Network

针对大型复杂网络难以控制的问题,本文主要对有向复杂网络的目标可控性进行研究。首先分析单输入下有向网络的目标控制,并将贪婪算法进行优化,同时对大型复杂网络的目标控制提出了新的算法。与贪婪算法相比,新算法引入对目标节点集施加免疫信号,可有效阻断免疫节点部分与外界的连接,有效避免干扰,提高网络目标控制的效率;新算法中的免疫方法结合随机免疫和熟人免疫这两种方法的优点于一体,可更大范围地移除网络中目标节点部分到非目标节点部分的出边。说明通过新算法寻找网络目标节点部分的最小驱动节点集更加高效方便。该研究为解决更为复杂图的目标可控性问题提供了方向和方法。

we mainly study the target control of complex networks. We first analyze the target control of directed net-works with single input. In addition, the greedy algorithm is optimized, and a new algorithm is proposed for the target control of large complex networks. Compared with the greedy algorithm, the new algorithm intro-duces the immune signal applied to the target node set, which can effectively block the connection of the im-mune nodes with the outside world, effectively avoid the interference and improve the efficiency of the network target control. The immune method in the new algorithm combines the advantages of both the random immuni-ty and the acquaintances immunity, and can remove the target node part of the network to the edge of the non-target node section in a wider range. The new algorithm is more efficient and convenient to find the minimum driver nodes set for the target control of complex networks. This study provides the direction and method for solving the problem of controllability of more complex graphs.