中圖分類號： TP391.1
文獻標識碼： A
DOI： 10.19358/j.issn.2096-5133.2022.01.003
引用格式： 黃冠杰. 基于Canopy-Kmeans算法的電力企業流量數據分析研究[J].信息技術與網絡安全，2022，41(1)：18-22.

Research on electric power enterprise flow data analysis based on Canopy-Kmeans algorithm

Abstract： Aiming at the phenomenon that a large number of business data of the key information infrastructure of electric power enterprises are vulnerable to network attacks, based on the existing network security equipment under each business information system, the flow data is collected through auxiliary equipment, and Canopy-Kmeans algorithm is used for data analysis and research. Firstly, through experiments, it is proved that the Canopy-Kmeans algorithm has a better clustering effect than the traditional K-means algorithm when processing flow data, and the accuracy rate is increased by about 11%. Then, the collected flow data of the power key business system is used，mining and analysis experiments are conducted based on the Canopy-Kmeans algorithm to complete the clustering of the same type of traffic data, analyze the characteristic items of attack traffic and business traffic, eliminate some misreporting information, and carry out network security protection work reasonably.

Key words : electricity；flow collection；Canopy-Kmeans；clustering；flow data analysis