Research On Dynamic State Estimation Theory Of Distribution Network Under Cyber Attacks

Real-time state estimation technology is critical for ensuring the safe operation of distribution networks.However,the growing openness of communication environments also creates new opportunities for network attacks,which can disrupt monitoring and control and threaten the security and stability of such networks.Therefore,this paper investigates the dynamic state estimation theory of distribution networks in the presence of network attacks,with a view to enhancing the network’s state estimation performance and improving its overall security.The main research objectives and contents are summarized as follows:1.The dynamic state estimation problem of distribution network under false data injection attacks is studied.To this end,FDI attacks on measurement and control signals are modeled as unknown inputs and outputs of system equations respectively.With the aim of improving the state estimation performance of distribution systems under such attacks,an improved adaptive singular value decomposition cubature Kalman filtering algorithm is designed by combining the cubature Kalman filtering algorithm and strong tracking algorithm.This algorithm enables accurate estimation of the distribution system state even when FDI attacks are present.2.The dynamic state estimation problem of distribution network under denial of service attacks is studied.Traditional state estimation algorithms that rely on data integrity might not work effectively due to transmission delays and persistent packet loss caused by Do S attacks.To address this issue,a new state estimation algorithm is proposed that enhances the accuracy of state estimation in distribution networks under Do S attacks.This algorithm considers the equivalent current flow measurement transformation technique to express the measurement equation as a linear model.Additionally,the phenomenon of partial measurement data delay and loss in the packet is explained using random variables that follow a Bernoulli distribution,which aids in reconstructing the dynamic model of the distribution network.Furthermore,an improved state estimator under Do S attacks is designed by combining the adaptive square-root cubature Kalman filter method,thereby enabling optimal state estimation of the system.3.The dynamic state estimation problem of distribution network under cyber attacks is studied.A new state estimation algorithm is proposed to improve the accuracy to obtain the optimal state estimation of distribution network against general cyber attacks.It is first proven that the optimal Kalman estimate can be decomposed into a weighted sum of local state estimates when Phasor Measurement Units being attacked.Focusing on the insecurity of the weighted sum method,a convex optimization based on local estimation is proposed to replace the method and combine the local estimation into a secure state estimation.Finally,it is proved that the proposed estimator is consistent with the Kalman estimator when all PMU measuring devices are functional.Additionally,a sufficient condition is established for ensuring estimator stability under attack when PMU devices experience abnormal measurements due to cyber attacks.