False Data Injection Attacks on Smart Grids: Attack Models, Challenges and Future Directions

Abstract Modernizing the electric power grid through advanced communications and intelligent devices has improved efficiency and reliability, but it has also expanded the cyberattack surface. One of the most consequential integrity threats is False Data Injection Attacks (FDIAs), which can compromise the measurements and data streams that underpin monitoring, protection, and control functions. This paper surveys FDIA research in smart grids and organizes attack targets into five domains: steady-state control; transient and auxiliary control; substation control; energy and load control; and AI-based systems. We review how attackers construct stealthy and data-driven FDIAs, summarize their physical and economic impacts, and discuss why many existing detection mechanisms remain vulnerable. Finally, we highlight open challenges and outline research directions toward resilient, attack-aware grid architectures.