新型电力系统中信息网络攻击行为辨识方法  

Method for Identifying Information Network Attack Behavior in New Power System

作　　者：任春梅 白欣雨 黄岩 REN Chunmei;BAI Xinyu;HUANG Yan(Xuchang KETOP Testing Research Institute Co.,Ltd.,Xuchang 461000,China)

机构地区：[1]许昌开普检测研究院股份有限公司,河南许昌461000

出　　处：《电工技术》2023年第13期233-237,共5页Electric Engineering

摘　　要：为提高新型电力系统信息网络的安全性和稳定性,需预防其中存在的攻击行为。为此,提出一种针对新型电力系统信息网络的攻击行为辨识方法。首先,分析攻击行为入侵信息网络的原理;其次,根据压缩感知理论对网络信息展开去噪处理;然后,利用去噪后的信息数据建立数据源矩阵及信息网络的样本协方差矩阵,并获取矩阵的最大特征值;最后,设定阈值函数,通过比较最大特征值与阈值函数来辨识新型电力系统信息网络中存在的攻击行为。实验结果表明该方法具有较高的辨识精度,能有效辨识出拒绝服务、端口扫描和协议异常三种攻击行为。In order to improve the security and stability of the new power system information network,it is necessary to prevent the existing attacks.Therefore,a method for identifying attack behavior of new power system information network is proposed.Firstly,the principle of attack behavior invading the information network was analyzed.Secondly,according to the compressed sensing theory,the network information was denoised.Thirdly,the denoising information data was used to establish the data source matrix and the sample covariance matrix of the information network,and the maximum feature value of the matrix was obtained.Finally,the threshold function was set,and the attack behavior in the new power system information network was identified by comparing the maximum feature value with the threshold function.Experimental results show that the proposed method has high identification accuracy and can effectively identify three attack behaviors:denial of service,port scanning and protocol abnormality.

关 键 词：新型电力系统 信息网络 压缩感知理论 样本协方差矩阵 攻击行为辨识 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]