This article presents a real-time non-probabilistic approach to detect load-redistribution (LR) attacks, which attempt to cause an overflow, in smart grids. Prior studies have shown that certain LR attacks can bypass traditional bad data detectors and remain undetectable, which implies that the presence of a reliable and intelligent detection mechanism is imperative. Therefore, in this study a detection mechanism is proposed based on the fundamental knowledge of the physics laws in electric grids. To do so, we leverage power systems domain insight to identify an underlying exploitable structure for the core problem of LR attacks, which enables the prediction of the attackers' behavior. Then, a fast greedy algorithm is presented to find the best attack vector and identify the most sensitive buses for critical transmission assets. Finally, a security index, which can be used in practice with minimal disruptions, is developed for each critical asset with respect to the identified best attack vector and sensitive buses. The proposed approach is applied to 2383-bus Polish test system to demonstrate the scalability and efficiency of the proposed algorithm.
- Cyber-attack detection
- false data injection attack (FDIA)
- greedy algorithm
- linear programming (LP)
- load-redistribution attack detection
ASJC Scopus subject areas
- Computer Science(all)