基于复杂网络的航班运行风险传播分析

为研究航班运行风险传播机理，基于民航航班运行程序，采集航班运行数据及航空公司、机场、机组、机务、空管等工作表现做为研究样本；采用经验建网法、时间序列相空间重构法和Spearman 相关系数3 种方法，构建面向航班运行风险传播问题的复杂网络，经过计算证实，使用Spearman 相关系数建网效果最佳；对应民航常用控制方法，引入重要度r 、改进感染率β′和改进恢复率γ′概念，提出适用于航班运行的改进SIR 模型，最后对风险网络传播进行动力学分析. 计算结果表明：重要度r =0.4 时，感染节点密度曲线峰值下降10%，达到峰值时间推迟5%；改进恢复率γ′=0.9 时，感染节点峰值降低6%. 证实加入重要度，改进恢复率可有效抑制风险网络传播；说明识别风险网络中对关键节点加以控制，提高风险节点恢复比率和速度，可有效提高航班安全保障能力.

Flight Operation Risk Propagation Based on Complex Network

In order to study the flight operations risk propagation mechanism, based on the civil aviation flight operation procedures, at first, the flight data and the working performances of airlines, airports, crew, aircraft and air traffic control were collected as research samples. Then, the experience network construction method and phasespace of time series reconstruction method and the Spearman correlation coefficient method were used to construct a complex network for flight operation risk propagation problems. After the calculation of the characteristic parameters, it was confirmed that the Spearman correlation coefficient method had built the network with best effect. After that, corresponding to the civil aviation common control methods, the concept of importance value r , improved infection rate β′ and improved recovery rate γ′ were introduced. An improved SIR model suitable for flight operation was proposed. Finally, the dynamic analysis of risk network propagation was carried out. The calculation results show that when the importance degree value equals 0.4, the peak value of the infection node density curve decreases by 10%, and the peak time is delayed by 5%. When the improved recovery rate is 0.9, the peak of the infected node is reduced by 6%. It is confirmed that adding the importance value and the improved recovery rate can effectively control risk networks spread. It shows that the identification and control to risk network key nodes, the improvement for the risk nodes recovery rate and speed, these can effectively improve the flight security.