摘要
传统的恶意数据检测从数学理论出发,通常以残差方程为基础,根据目标函数的偏离进行检测。虚假数据注入攻击(FDIA)通过构造与雅克比矩阵列向量线性相关的攻击矢量,针对电力系统状态估计发起蓄意攻击。理论上FDIA躲过了电力系统的恶意数据检测机制,使原方法对于FDIA失效。结合CPS分析了恶意数据检测的原理,以及假数据注入攻击的原理和方式。在IEEE 30和IEEE 118节点系统上,通过仿真实验的方法,对FDIA使用标准残差检测法和目标函数极值法进行检测。实验结果证明了传统的不良检测对假FDIA的局限性。
The traditional malicious data detection starts from the mathematical theory,usually based on the residual equation and based on the deviation of the objective function.False data injection attack(FDIA)is a deliberate attack on power system state estimation by constructing attack vectors linearly related to Jacobian matrix column vectors.Theoretically,FDIA can evade the malicious data detection mechanism of the power system and make the original method invalid for FDIA.Based on CPS,this paper analyzes the principle of malicious data detection and the principle and method of FDIA.On the IEEE 30 and IEEE 118 node systems,the attack on the false data injection is detected by using the standard residual detection method and the objective function extremum method through the simulation experiment.The experimental results prove the limitation of the traditional bad detection on the FDIA.
作者
武津园
王勇
刘丽丽
程彦喆
WU Jinyuan;WANG Yong;LIU Lili;CHENG Yanzhe(School of Computer Science and Technology,Shanghai University of Electric Power,Shanghai 200090,China;National Energy Distributed Energy Technology Research and Development(Experimental)Center,Huadian Electric Power Research Institute Co.,Ltd.,Hangzhou 310030,China;Cyberspace Security College,Southeast University,Nanjing 210000,China)
出处
《上海电力大学学报》
CAS
2020年第6期591-597,共7页
Journal of Shanghai University of Electric Power
基金
国家自然科学基金(61772327)
上海自然科学基金(16ZR14366300)
浙江大学工业控制技术国家重点实验室开放式基金(ICT1800380)
上海电力学院智能电网产学研开发中心基金(A-0009-17-002-05)
奇安信大数据协同安全国家工程实验室开放课题(QAX-201803)。
关键词
智能电网
假数据注入
状态估计
smart grids
false data injection attack
state estimation
