Abstract
This paper demonstrates that false data injection (FDI) attacks are extremely limited in their ability to cause physical consequences on $N-1$ reliable power systems operating with real-time contingency analysis (RTCA) and security constrained economic dispatch (SCED). Prior work has shown that FDI attacks can be designed via an attacker-defender bi-level linear program (ADBLP) to cause physical overflows after re-dispatch using DCOPF. In this paper, it is shown that attacks designed using DCOPF fail to cause overflows on $N-1$ reliable systems because the system response modeled is inaccurate. An ADBLP that accurately models the system response is proposed to find the worst-case physical consequences, thereby modeling a strong attacker with system level knowledge. Simulation results on the synthetic Texas system with 2000 buses show that even with the new enhanced attacks, for systems operated conservatively due to $N-1$ constraints, the designed attacks only lead to post-contingency overflows. Moreover, the attacker must control a large portion of measurements and physically create a contingency in the system to cause consequences. Therefore, it is conceivable but requires an extremely sophisticated attacker to cause physical consequences on $N-1$ reliable power systems operated with RTCA and SCED.
Introduction: The "Safety Buffer" of the Grid
For years, the cybersecurity community has warned about False Data Injection (FDI) attacks, where hackers manipulate SCADA measurements to trick grid operators into seeing a "fake" system state. The fear is that operators, acting on this false data, would re-dispatch power in a way that causes real physical overflows and equipment damage. However, most previous studies used simplified models like DC Optimal Power Flow (DCOPF). This paper investigates whether the modern $N-1$ reliability criterion—the requirement that the grid must remain stable even if any single component fails—acts as a natural defense against such attacks.
Methodology: Modeling the Strong Attacker
The researchers compared two types of attackers: a "Standard" attacker using a basic DCOPF model and a "Strong" attacker with perfect knowledge of the grid’s internal algorithms, including Real-Time Contingency Analysis (RTCA) and Security Constrained Economic Dispatch (SCED). To find the most damaging attacks, they utilized an Attacker-Defender Bi-Level Linear Program (ADBLP) on the Synthetic Texas System, a massive benchmark model featuring 2,000 buses and over 3,000 branches.
As shown in the flowchart above, the attack is a complex interaction where the system's automated response (RTCA and SCED) can effectively mitigate the intended damage of the injected data. The study implemented these scenarios on the OpenPA platform to mimic real-world Energy Management Systems (EMS).
Key Insights and Results
The study revealed that the industry-standard $N-1$ reliability constraints create a significant "slack" in the system that absorbs the impact of FDI attacks.
- The DCOPF Illusion: Attacks designed using simplified DCOPF models resulted in 0% physical overflow when tested against a realistic $N-1$ system. The grid’s standard operating procedures simply absorbed the "fake" load shifts.
- Conservative Operation as a Shield: Because $N-1$ standards force the grid to operate conservatively, an FDI attack rarely manages to push the system into a truly dangerous state during normal operation.
- Post-Contingency Vulnerability: Even a "Strong" attacker struggled to cause damage during normal operation. Physical consequences typically only manifested after a contingency occurred (e.g., a line failure).
Figure 4 illustrates this discrepancy. While the attacker "predicts" a significant overflow (red line), the actual physical flow (blue dashed line) often remains within safe limits or only slightly exceeds them under specific conditions. The grid's internal SCED keeps the "cyber" view (green line) within limits, effectively masking the attack's physical intent.
Conclusion
The $N-1$ reliability criterion significantly hardens the power grid against FDI attacks. To cause meaningful physical damage, an attacker would need to control a staggering number of sensors (nearly half the system in some simulations) and likely coordinate a physical failure alongside the cyber-attack. While theoretically possible, the sophistication required makes the grid much more resilient than simplified academic models suggest.
Summary updated