Fusion State Estimation for Power Systems under DoS Attacks: A Switched System Approach

Ji Chen,Chunxia Dou,Li Xiao,Zhuo Wang

DOI: https://doi.org/10.1109/tsmc.2019.2895912

2019-01-01

IEEE Transactions on Systems Man and Cybernetics Systems

Abstract:This paper proposes a switched system method (SSM) for the fusion state estimation (SE) under denial-of-service (DoS) attacks in power systems. One of the subsystems in the SSM employs the widely used dynamic model with unit state transition matrix and zero control input. The performance of this model is favorable when the system operates in normal conditions but tends to suffer while the system state dramatically changes. Moreover, its accuracy and convergence is sensitive to the value of the process noise covariance matrix. The other subsystem, with some opposite characteristics, utilizes a pseudo-dynamic model derived from the linearization of the power flow equation. A switching rule based on the innovations is established to make a tradeoff between the two subsystems with regard to estimation accuracy, converge speed, and computing time. Apart from the difference in the reporting rate of various measurements, cyber attacks, such as DoS attacks may result in the absence of measurements in the snapshot of SE at the finest time scale. To deal with this problem, the linear extrapolation and multi-step prediction techniques are involved. Simulation studies are conducted on the IEEE-14 bus test system and a 33 bus distribution system under various scenarios, and the performance of the proposed method is discussed.