工业变电站运维系统异常数据入侵检测互信息实现  

Mutual Information Implementation for Anomaly Data Intrusion Detection in Industrial Substation Operation and Maintenance System

作　　者：王子杰 潘啸天 Wang Zijie;Pan Xiaotian(Jurong Power Supply Branch,State Grid Jiangsu Power Co.,Ltd.,Jurong Jiangsu 212400,China)

机构地区：[1]国网江苏省电力有限公司句容市供电分公司,江苏句容212400

出　　处：《现代工业经济和信息化》2024年第8期94-95,104,共3页Modern Industrial Economy and Informationization

摘　　要：工业变电站运维系统在运行的过程中经常受到不同类型的数据侵入,严重影响到变电站的安全,进而造成很大的经济损失。为了进一步提高运维系统的安全,设计了一种面向互信息技术的工业变电站运维系统异常数据入侵检测方法,并开展测试分析,证明了本文方法的准确性。研究结果表明:相对于PCA算法,互信息(MI)算法获得更高的特征提取精度,检测率也明显提升,降低了误报率。当数据量快速增加后,分布式模型表现出了更短的入侵检测时间。该研究对提高运维系统异常数据入侵检测稳定性具有一定的实践指导意义,但在小概率攻击类型中该算法存在导致检测率为零结果,期待后续进一步的加强。Industrial substation operation and maintenance system is often subject to different types of data intrusion in the process of operation,which seriously affects the safety of the substation and thus causes great economic losses.In order to further improve the safety of the operation and maintenance system,a mutual information technology oriented method for detecting abnormal data intrusion in industrial substation operation and maintenance system is designed,and test analyses are carried out to prove the accuracy of the method in this paper.The research results show that compared with the PCA algorithm,the mutual information(MI)algorithm obtains higher feature extraction accuracy,and the detection rate is also significantly improved,reducing the false alarm rate.The distributed model exhibits shorter intrusion detection time when the data volume increases rapidly.The study has certain practical guidance for improving the stability of abnormal data intrusion detection in operation and maintenance systems,but the algorithm exists in small probability attack types leading to zero detection rate results,and further enhancement is expected in the follow-up.

关 键 词：工业变电站 运维系统 互信息法 入侵检测 

分 类 号：TH17[机械工程—机械制造及自动化]