| 引用本文: | 杨奕贤,郭力,王洪达,李霞林,张涛,黄生俊,朱想,王成山.基于数据驱动的直流微电网虚假数据注入攻击快速防御策略[J].电力自动化设备,2021,41(5): |
| YANG Yixian,GUO Li,WANG Hongda,LI Xialin,ZHANG Tao,HUANG Shengjun,ZHU Xiang,WANG Chengshan.Fast defense strategy of false data injection attack in DC microgrid based on data-driven[J].Electric Power Automation Equipment,2021,41(5): |
|
| 本文已被：浏览 4267次 下载 1660次 |
|
|
|
| 基于数据驱动的直流微电网虚假数据注入攻击快速防御策略 |
|
杨奕贤1, 郭力1, 王洪达1,2, 李霞林1, 张涛3, 黄生俊3, 朱想4, 王成山1
|
|
1.天津大学 智能电网教育部重点实验室，天津 300072;2.海军勤务学院 供应管理系，天津 300450;3.国防科技大学 系统工程学院 多能源系统智慧互联技术湖南省重点实验室，湖南 长沙 410073;4.中国电力科学研究院(南京)，江苏 南京 210003
|
|
| 摘要: |
| 基于分布式控制的直流微电网是一种典型的信息物理系统，攻击者可在信息层利用虚假数据注入攻击(FDIA)的方式使微电网偏离运行目标，从而影响微电网的供电质量。为此，分析了虚假数据的作用途径，对虚假数据的不利影响进行了机理性解释与建模，提出了一种基于数据驱动的FDIA快速防御策略。通过离线学习、在线判断直流微电网的暂态扰动过程，分辨出扰动的来源是正常负荷变化还是虚假数据，然后通过降低虚假数据权重的方式将网络攻击产生的不利影响降至最低。算例仿真结果验证了所提防御策略的有效性。 |
| 关键词: 直流微电网 分布式控制 信息物理系统 虚假数据注入攻击 网络攻击 防御策略 数据驱动 |
| DOI：10.16081/j.epae.202105022 |
| 分类号:TM761 |
| 基金项目:国家自然科学基金资助项目(51977142)；天津市自然科学基金资助项目(20JCQNJC00350) |
|
| Fast defense strategy of false data injection attack in DC microgrid based on data-driven |
|
YANG Yixian1, GUO Li1, WANG Hongda1,2, LI Xialin1, ZHANG Tao3, HUANG Shengjun3, ZHU Xiang4, WANG Chengshan1
|
|
1.Key Laboratory of Smart Grid of Ministry of Education, Tianjin University, Tianjin 300072, China;2.Department of Supply Management, Naval Logistics Academy, Tianjin 300450, China;3.Hunan Key Laboratory of Multi-energy System Intelligent Interconnection Technology, College of Systems Engineering, National University of Defense Technology, Changsha 410073, China;4.China Electric Power Research Institute(Nanjing),Nanjing 210003, China
|
| Abstract: |
| DC microgrid based on distributed control is a typical cyber-physical system. In the cyber layer, the attacker can make the microgrid deviate from the operation target through FDIA(False Data Injection Attack) way, thus affecting the power supply quality of microgrid. Therefore, the function chennel of false data is analyzed, the adverse effects of false data are explained and modeled mechanically, and a fast defense strategy of FDIA based on data-driven is proposed. Through off-line learning and on-line judgment of DC microgrid’s transient disturbance process, the source of disturbance can be distinguished whether it is form normal load change or false data. Then, the adverse effects of cyber attack can be minimized by reducing the weight of false data. Simulative results of a numerical example verify the effectiveness of the proposed defense strategy. |
| Key words: DC microgrid distributed control cyber-physical system false data injection attack cyber attack defense strategy data-driven |
