基于随机矩阵的电网虚假数据注入攻击策略

ATTACK STRATEGY OF FALSE DATA INJECTION IN POWER GRID BASED ON RANDOM MATRIX

  • 摘要: 在有限的时间窗口访问测量值条件下,提出一种基于随机矩阵的电网虚假数据注入攻击策略。利用多个冗余传感器测量某过滤噪声的影响,并获得对系统状态的最佳估计;引入一种改进随机矩阵算法来构建虚假数据注入攻击向量,以应对无法以高概率统计数据检测器的有限测量,进一步描述攻击的坏数据检测器绕过概率与其稀疏性之间的重要权衡,从而有效保证了攻击的空间相关性;通过IEEE总线系统仿真验证了该方法能够增强绕过坏数据检测器的概率,增强攻击效果。

     

    Abstract: Under the condition of limited time window to access the measured values, a false data injection attack strategy in power grid based on random matrix is proposed. Multiple redundant sensor measurements were used to filter the influence of noise, and the best estimation of system state was obtained. An improved random matrix algorithm was introduced to construct false data injection attack vector to deal with the limited measurement that could not bypass the bad data detector with high probability. The important trade-off between the bypass probability of the bad data detector and its sparsity were described, so as to effectively ensure the spatial correlation of the attack. The IEEE bus system simulation verifies that the proposed method can enhance the probability of bypassing the bad data detector and enhance the attack effect.

     

/

返回文章
返回