| 摘要: |
| 随着分布式电源在电网中所占比重的不断提升，针对分布式电源的攻击将给电网带来更严重的安全威胁。攻击者可以通过网络入侵手段协同控制电网中防御较弱的配网侧分布式电源功率输出，最终影响发电侧发电机等关键设备的安全运行。为保障电网安全稳定运行，亟需研究针对分布式电源接入场景下的安全威胁及其防御措施。首先，本文在电力系统动态模型基础之上建立了电网振荡攻击的最小代价攻击模型，通过协同控制多个分布式电源的功率，在牺牲最少被控节点的前提下导致电网发生振荡。其次，针对现有振荡检测算法的不足，本文提出一种启发式的攻击源检测算法，通过分析系统内各节点的势能变化，可有效辅助定位攻击源。算例仿真分析结果验证了通过最小代价攻击影响电网稳定运行的可行性，以及攻击检测方法的有效性。 |
| 关键词: 智能电网 分布式电源 信息-物理攻击 振荡攻击 攻击检测 |
| DOI：10.19363/J.cnki.cn10-1380/tn.2019.05.05 |
| 投稿时间：2019-01-28修订日期：2019-04-18 |
| 基金项目:本课题得到国家重点研发计划资助项目（No.2016YFB0800202）；国家自然科学基金（No.61772408，No.U1766215，No.U1736205，No.61721002，No.61632015，No.61833015）；霍英东基金（No.151067）和中央高校基本科研业务费专项资金资助。 |
|
| Oscillation Attack Modelling and Detection with Penetration of Distributed Energy Resources in Smart Grid |
| LIU Yang,GUI Yuhong,AN Dou,GUAN Xiaohong,LIU Ting |
| Ministry of Education Key Lab for Intelligent Networks and Network Security, Xi'an Jiaotong University, Xi'an 710049, China |
| Abstract: |
| The increasing distributed renewable resources (DERs) will introduce more vulnerabilities to smart grids. In this scenario, attackers can change the power output of DERs in the distribution network, and finally disturb the secure operation of crucial devices such as generators in the transmission network. To secure the operation safety and stability in smart grids, it is necessary to study the threats and relevant defense strategies under the scenario with high penetration of DERs. First, a novel attack model is proposed based on power system dynamics model, which could can cause power system oscillation with minimal attack costs via coordinated control of DERs maliciously. Second, in consideration of current oscillation detection algorithms' limitations, a heuristic attack detection method is proposed. The simulation-based case studies demonstrate that the proposed attack could threat power system operational safety with minimal costs, and the detection method can effectively locate the real attack sources. |
| Key words: smart grid distributed energy resources (DERs) cyber-physical attack oscillation attack attack detection |
