基于自适应加权混合预测的电网虚假数据注入攻击检测  

Grid False Data Injection Attack Detection Based on Adaptive Weighted Hybrid Prediction

作　　者：束洪春 杨永银[2,3] 赵红芳 许畅 赵学专[2,3] SHU Hongchun;YANG Yongyin;ZHAO Hongfang;XU Chang;ZHAO Xuezhuan(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,Yunnan Province,China;Yunnan Key Laboratory of Green Energy,Electric Power Measurement Digitalization,Control and Protection(Kunming University of Science and Technology),Kunming 650500,Yunnan Province,China;Faculty of Electric Power Engineering,Kunming University of Science and Technology,Kunming 650500,Yunnan Province,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南省昆明市650500 [2]昆明理工大学(云南省绿色能源与数字电力量测及控保重点实验室),云南省昆明市650500 [3]昆明理工大学电力工程学院,云南省昆明市650500

出　　处：《电网技术》2025年第3期1246-1256,I0095,共12页Power System Technology

基　　金：国家自然科学基金重点项目(52037003);云南省重大科技专项计划项目(202002AF080001)。

摘　　要：电力系统作为实时信息与能源高度融合的电力信息物理融合系统(cyber-physical power system,CPPS),虚假数据注入攻击(false data injection attacks,FDIAs)的准确辨识将有效保证CPPS安全稳定运行。为准确、高效地完成日前负荷预测,首先使用肯德尔相关系数(Kendall's tau-b)量化日期类型的取值,引入加权灰色关联分析选取相似日,再建立基于最小二乘支持向量机(least squares support vector machine,LSSVM)的日前负荷预测模型。将预测负荷通过潮流计算求解的系统节点状态量与无迹卡尔曼滤波(unscented Kalman filter,UKF)动态状态估计得到的状态量进行自适应加权混合,最后基于混合预测值和静态估计值间的偏差变量提出了攻击检测指数(attack detection index,ADI),根据ADI的分布检测FDIAs。若检测到FDIAs,使用混合预测状态量对该时刻的量测量进行修正。使用IEEE-14和IEEE-39节点系统进行仿真,结果验证了所提方法的有效性与可行性。As a cyber-physical power system(CPPS)that integrates real-time information and energy,accurate identification of false data injection attacks(FDIAs)is essential to ensure the secure and stable operation of the CPPS.To achieve accurate and efficient day-ahead load forecasting,Kendall's tau-b coefficient is first used to quantify the values of data types,and weighted grey relational analysis is introduced to select similar days.Then,a day-ahead load forecasting model based on the least squares support vector machine(LSSVM)is established.The predicted load is mixed adaptively with the system's state variables obtained from power flow calculation and unscented Kalman filter(UKF)dynamic state estimation.Finally,an attack detection index(ADI)is proposed based on the deviation between the mixed prediction values and static estimation values.FDIAs are detected based on the distribution of ADI.If FDIAs are detected,the mixed prediction state variables are used to correct the measured values at that time.The effectiveness and feasibility of the proposed method are verified through simulations on the IEEE-14 and IEEE-39 node systems.

关 键 词：电力信息物理系统 加权灰色关联分析 无迹卡尔曼滤波 最小二乘支持向量机 虚假数据攻击 攻击检测指数 

分 类 号：TM721[电气工程—电力系统及自动化]