基于DeepWalk算法的电力系统错误数据注入网络攻击分类方法  

Cyber attack classification method of false data injection in power system based on DeepWalk algorithm

作　　者：连祥龙 钱瞳 张银 唐文虎[1] LIAN Xiangong;QIAN Tong;ZHANG Yin;TANG Wenhu(School of Electric Power Engineering,South China University of Technology,Guangzhou 510641,China)

机构地区：[1]华南理工大学电力学院,广东广州510641

出　　处：《电力自动化设备》2023年第3期166-171,共6页Electric Power Automation Equipment

基　　金：国家自然科学基金资助项目(51977082);中国博士后科学基金资助项目(2021M701239)。

摘　　要：为了准确、有效地识别错误数据注入攻击(FDIA)对电网造成危害的严重程度,提出了基于DeepWalk算法的FDIA分类新方法。根据FDIA的特点,构建电力系统的响应模型;提出批量随机边删减策略,将响应模型生成的攻击数据构造为攻击场景图;采用DeepWalk算法将攻击场景图中的节点映射为低维向量,并将其作为机器学习算法的输入对FDIA进行分类。以遭受FDIA的IEEE 39节点系统为例进行仿真,结果表明所提方法可以根据FDIA对电网造成危害的严重程度准确、有效地对FDIA进行分类。In order to accurately and effectively identify the severity of false data injection attacks(FDIAs) on power grid,a novel FDIA classification method based on DeepWalk algorithm is proposed. According to the characteristics of FDIAs,the response model of power system is constructed. The batch random edge reduction strategy is proposed to construct the attack data generated by the response model as the attack scenario graph. The DeepWalk algorithm is used to map the nodes in the attack scenario graph into lowdimensional vectors,which are used as the inputs of the machine learning algorithm to classify the FDIAs.The simulative results of the IEEE 39-bus system suffering from FDIAs show that the proposed method can accurately and effectively classify FDIAs according to the severity of the damage caused by FDIAs to the power grid.

关 键 词：电力系统 网络攻击 错误数据注入攻击 DeepWalk算法 节点分类 

分 类 号：TM761[电气工程—电力系统及自动化]