Secure Dynamic State Estimation of WECS-Based Networked Microgrids Against Historical Measurement Triggered DoS Attacks

Wind energy conversion systems (WECSs) based networked microgrids has been widely used in recent years. The mean square error (MSE) metric can yield imprecise outcomes if measurement data is polluted by non-Gaussian disturbances or extreme values. To address this problem, we propose a new robust square root cubature Kalman filter (SRCKF) method called maximum correlation criterion (MCC)-SRCKF, which incorporates MCC into the SRCKF framework of dynamic state estimation. In MCC, by considering the high-order moments of the error distribution, it demonstrates anti-interference ability against non-Gaussian noise, thus serving as an ideal alternative in the MSE cost function field of SRCKF. Furthermore, within the framework of SRCKF, this study introduces statistical linear regression models and non-moving point iteration strategies to solve the optimal state estimation under MCC conditions. Therefore, a historical measurement triggered DoS attack model is proposed from the attacker's perspective, aiming to destabilise the WECS-based networked microgrids. The security conditions of the power system under such attacks are obtained. The proposed method is validated numerically using an IEEE 39-bus system, and the results demonstrate its effectiveness and superiority.