Small-world network is a very common network structure, which is characterized by low average degree, small average path length and high centrality. At the same time, small-world networks have high resilience to random errors and low resilience to targeted attacks. In this study, the importance of nodes is represented by attributes such as degree and centrality, and attacks refer to the removal of important nodes. The network is attacked according to the degree, betweenness and closeness centrality to observe the power distribution. The data is mainly obtained from the open source OpenFlight. Gephi, Python, and Excel are used as tools. Gephi is used for network visualization and analysis. The third-party python libraries Pandas, Matplotlib, and NetworkX were used in this study to deal with the things that Gephi can't compute or represent well, and then plot the corresponding graphs with Matplotlib. The work of cleaning data is mainly done by excel.