Enhancing Cyber-Security of Distributed Robust State Estimation: Identification of Data Integrity Attacks in Multi-Operator Power System

Abstract

State estimation (SE) has a crucial role to play in the monitoring and control of power grids. Although currently the SE is typically done in a centralized or hierarchical manner, distributed SE will become a significant alternative to centralized and hierarchical approaches in the future smart grids. This is because the power grids will be increasingly interconnected in future smart grids and the complexity scale of an interconnection will render centralized SE computationally formidable. Performing distributed SE requires leveraging advanced communication and computation technology. Nevertheless, relying on communication networks raises its susceptibility to data integrity attacks, such as false data injection (FDI) attacks. In this paper, we demonstrate that the attacker who compromises the communication infrastructure can launch an FDI attack on distributed SE which could circumvent present robust estimators and bad data detectors. Afterwards, to effectively defense against the proposed FDI attack, two detection methods are proposed for two different modes of an interconnected power system. A detector is developed that validates the error of estimates of the state variables relative to their actual value as an index using a threshold value for different areas when the network is being run by an operator. A controlled information dissemination strategy is utilized to securely notify all areas of each other's proposed index when the network is being run by multiple operators. The proposed algorithms are validated on the IEEE 14-bus test system.