A Review of Detection, Evolution, and Data Reconstruction Strategies for False Data Injection Attacks in Power Cyber-Physical Systems

Abstract

The integration of information and physical systems in modern power grids has heightened vulnerabilities to False Data Injection Attacks (FDIAs), threatening the secure operation of power cyber-physical systems (CPS). This paper reviews FDIA detection, evolution, and data reconstruction strategies, highlighting cross-domain coordination, multi-temporal evolution, and stealth characteristics. Challenges in existing detection methods, including poor interpretability and data imbalance, are discussed, alongside advanced state-aware and action-control data reconstruction techniques. Key issues, such as modeling FDIA evolution and distinguishing malicious data from regular faults, are identified. Future directions to enhance system resilience and detection accuracy are proposed, contributing to the secure operation of power CPS.