| 基于TNPE的智能电网虚假数据注入攻击检测* |
| |
| 引用本文： | 曾俊娆,李鹏,高莲,沈鑫.基于TNPE的智能电网虚假数据注入攻击检测*[J].中国安全生产科学技术,2021,17(3):124-129. |
| |
| 作者姓名： | 曾俊娆 李鹏 高莲 沈鑫 |
| |
| 作者单位： | （1.云南大学 信息学院，云南 昆明 650500； 2.云南省高校物联网技术及应用重点实验室，云南 昆明 650500;3.云南电网有限责任公司 电力科学研究院，云南 昆明 650511) |
| |
| 基金项目： | 国家自然科学基金项目(61763049);云南省应用基础研究重点课题项目(2018FA032)。 |
| |
| 摘 要： | 为实现智能电网中虚假数据注入攻击的实时检测，提高电力系统运行的安全性，采用1种基于时序近邻保持嵌入的方法，对正常状态下采集到的电网历史量测数据建立离线模型，得到T2统计限，将实时数据通过模型获得的T2统计量与离线模型的统计限进行对比，若超过统计限，则说明存在虚假数据注入攻击。该方法在提取局部空间结构特征的基础上，可同时获得与时间相关的动态特征。在IEEE30节点测试系统上进行仿真实验，并与ICA，PCA，NPE方法进行比较。结果表明:所提方法有高达100%的检测率，且有较低的误报率，能够有效应用在虚假数据注入攻击的检测中。
|
| 关 键 词： | 电力系统 状态估计 时序近邻保持嵌入 攻击检测 |
Detection of false data injection attacks in smart grids based on time neighbor preserving embedding(TNPE) |
| |
| ZENG Junrao,LI Peng,GAO Lian,SHEN Xin.Detection of false data injection attacks in smart grids based on time neighbor preserving embedding(TNPE)[J].Journal of Safety Science and Technology,2021,17(3):124-129. |
| |
| Authors: | ZENG Junrao LI Peng GAO Lian SHEN Xin |
| |
| Affiliation: | (1.School of Information,Yunnan University,Kunming Yunnan 650500,China;2.Internet of Things Technology and Application Key Laboratory of Universities in Yunnan,Kunming Yunnan 650500,China;3.Yunnan Power Grid Co.,Ltd.,Kunming Yunnan 650511,China) |
| |
| Abstract: | In order to realize the real-time detection of false data injection attacks in the smart grids and improve the safety of power system operation,a method based on the time neighbor preserving embedding(TNPE)was adopted,and an offline model was established for the historical measurement data of the power grid collected under normal conditions to obtain T2 statistical limit.The T2 statistics obtained through the model by the real-time data was compared with the T2 statistical limit of the offline model,and if the statistical limit was exceeded,it indicated that there was a false data injection attack.On the basis of extracting the local spatial structure features,this method could simultaneously obtain the time-related dynamic features.The simulation experiment was carried out on the IEEE30 node test system and compared with the ICA,PCA,and NPE methods.The results showed that the method proposed in this paper had a detection rate of up to 100%and a low false alarm rate,which can be effectively applied in the detection of false data injection attacks. |
| |
| Keywords: | power system state estimation time neighbor preserving embedding attack detection |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《中国安全生产科学技术》浏览原始摘要信息 |
|
点击此处可从《中国安全生产科学技术》下载全文 |
