基于改进K-shell算法的空中交通信息物理系统节点排序

Node ranking of air traffic information physical system based on improved K-shell algorithm

为精确识别空中交通信息物理系统(cyber physical system,CPS)节点影响力,依据空中交通管理系统信息网络和物理网络的深度耦合关系,结合复杂网络理论,构建空中交通CPS,并对原有K-shell算法进行改进,重新定义了加权度指标。通过改进K-shell算法分别对华东空中交通CPS信息网的管制席位与物理网的航路点影响力进行排序,同时与度、度中心性、介数中心性、接近中心性、特征向量中心性、K-shell等排序方法进行对比,证明了改进K-shell算法能够有效识别网络中节点的影响力,特别是对航路网这种无标度航空网络的节点影响力识别,改进K-shell算法计算结果比其他方法更为精确。最后,对空中交通CPS信息网与物理网的影响力进行分析,证明了信息网中影响力大的管制席位管理的扇区,所辖航路点影响力也偏大,因此应保护影响力大的节点,尽可能避免节点失效引发空中交通CPS网络大面积瘫痪,减少航班延误的发生。

In order to accurately identify the influence of air traffic cyber physical system(CPS)nodes,based on the deep coupling relationship between the information network and the physical network in the air traffic management system,combined with complex network theory,an air traffic CPS network model was constructed.The original K-shell algorithm was improved,and the weighting index was redefined.The improved K-shell algorithm was used to rank the control seats of the East China air traffic CPS information network and the airway points of the physical network.At the same time,comparisons were made with ranking methods such as degree,degree centrality,betweenness centrality,closeness centrality,eigenvector centrality,and K-shell,which proved that the improved K-shell algorithm can effectively identify the influence of nodes in the network.The improved K-shell algorithm was more accurate than other methods especially for the node influence recognition of scale-free aviation network such as airway network.Finally,the influence of the air traffic CPS information network and the physical network were analyzed.It is proved that the effect of airway points within the sector with large influence in the information network are also large.The nodes with great influence should be protected to avoid large area paralysis of air traffic CPS network caused by node failure as far as possible,so as to reduce the occurrence of flight delay.