Study On The Robustness Of The Usairline Network Based On Complex Networks

Based on the complex networks theory, this thesis takes the US Air line network as the main research object, takes the robustness of complex networks as research target. With the two most common types of typical complex network model (SW small world network and BA scale-free network) for comparison, according to the static and dynamic conditions, we get the simulation results of different networks and analysis them.With Complex network analysis software pajek, we carry on the empirical research of the US Air line network. The empirical results imply that the degree distribution of the US Air line network follows a power law, p(k)∝k-r, degree exponent r≈1.8. So it is a scale-free network. In addition, the Average path length of the US Air line network is2.74, obviously the US Air line network gets Small world network characteristics.In the static condition, through the different attack modes (random perturbations and intentional attacks), we conduct the experiment for getting the robustness results of SW small world network, BA scale-free network and the US Air line network. The results show that under random perturbations, the robustness between BA scale-free network and the US Air line network is higher than SW small world network; under intentional attacks, the robustness of SW small world network is the highest, There is not much difference between BA scale-free network and the US Air line network, precisely speaking, the robustness of the US Air line network is smaller than BA scale-free network.In the dynamic condition, we propose a new cascading failure model, Adopting the initial load of a node i to be Lj=(BjKj)α with Ki and Bi being the degree and the betweenness of the node i, respectively, where a is a tunable parameter and governs the strength of the initial load of a node. We conduct cascading failure simulation research on the US Air line network, SW small world network and BA scale-free network. The results show that SW small world network have a threshold value a,when parameter a is smaller than ac, cascading failure won’t happen, otherwise, it happen. Threshold value α,don’t exist between the US Air line network is smaller than BA scale-free network, the robustness of SW small world network is higher than the US Air line network, the robustness of the US Air line. network is higher than BA scale-free network. Another important discovery is that in BA scale-free network, although it is feasible through Spending more cost to increase the robustness, when a is more, that’s impossible.