Mitigation of Cyber-Threat on Dynamic State Estimation

Dynamic state estimation plays a significant role in raising the situational awareness of cyber-physical energy systems. Determining the dynamic states of the system is dependent on the type of measurement data of the power grid and the tools used for the estimation such as Kalman filters (KF). However, due to the modernization of the power grid and the various tools for data acquisition, the grid has become vulnerable to new types of threats such as false data injections (FDIs) attacks. In this work, a new approach is presented to mitigate the impact of FDIs on dynamic state estimation. We use an unscented Kalman filter aided by eXtreme Gradient Boosting (XGB) classifier as a two-stage approach to mitigate the impact of FDIs and cyber-threats. The proposed approach is tested on the New England IEEE 39-Bus System under different operating conditions. The results show the effectiveness of the proposed approach in mitigating the impact of UK-XGB on dynamic state estimation.