山东电力技术2024，Vol.51Issue(3)：18-26,9.DOI:10.20097/j.cnki.issn1007-9904.2024.03.003
基于对抗性自动编码器的城市配电网虚假数据注入攻击检测
Detection of False Data Injection Attack in Urban Distribution Network Based on Adversarial Autoencoder
摘要
Abstract
Under the background which considerably integrates information and physics,fast and accurate detection of false data injection attacks(FDIAs)is the key to the safe and stable operation of urban distribution network.FDIAs detection method for urban distribution network was proposed based on an adversarial autoencoder.We combined the autoencoder with the generative adversarial network,which can extract data features,find anomalies in distribution network data caused by FDIAs,and utilize a small amount of labeled data for network training to avoid high labeling costs.It also reduces the dependence of FDIAs detection on the network topology.The simulation and analysis results of the typical distribution network cases demonstrate that the proposed method is suitable for the ever-growing urban distribution networks,which indicates its advantages over the current FDIAs detection methods in accuracy and efficiency.关键词
城市配电网/虚假数据注入攻击检测/生成对抗网络/半监督Key words
urban distribution network/false data injection attack detection/generative adversarial network/semi-supervised分类
信息技术与安全科学引用本文复制引用
常颢,徐俊俊,王晓兵,周宪..基于对抗性自动编码器的城市配电网虚假数据注入攻击检测[J].山东电力技术,2024,51(3):18-26,9.基金项目
国家自然科学基金资助项目(52107101) （52107101）
江苏省自然科学基金资助项目(BK20200761).National Natural Science Foundation of China(52107101) （BK20200761）
Natural Science Foundation of Jiangsu Province(BK20200761). （BK20200761）
