Deceiving Flexibility: A Stealthy False Data Injection Model in Vehicle-to-Grid Coordination

A team of researchers has identified a critical cybersecurity vulnerability in the systems that manage electric vehicles (EVs) as grid resources. In a paper titled "Deceiving Flexibility: A Stealthy False Data Injection Model in Vehicle-to-Grid Coordination," authors Kaan T. Gun, Xiaozhe Wang, and Danial Jafarigiv demonstrate a novel attack that targets the centralized coordination models, like the extended State Space Model (eSSM), used to scale V2G systems. Unlike attacks that physically disrupt charging, this method involves an adversary compromising only a subset of connected EVs to subtly falsify the data they report back to the grid operator.

The attack manipulates two key data points: the State of Charge (SoC) and the power measurements of the vehicles. By carefully crafting this false data, the attacker can deceive the grid operator's perception of how much flexible energy capacity the entire EV fleet can provide. Crucially, the false data is designed to remain consistent with the grid's model-based expectations, allowing it to evade standard anomaly detection systems. Numerical simulations confirm that this stealthy False Data Injection Attack (FDIA) can successfully deteriorate grid frequency stability, posing a risk to overall power reliability, all without the attacker needing direct access to physical control infrastructure.

This research shifts the focus from physical disruption to data integrity attacks within aggregated energy frameworks. The findings expose a significant blind spot in current V2G security models, which may not be designed to detect coordinated data deception from a minority of compromised assets. The paper concludes by emphasizing the urgent need for enhanced, tailored detection and mitigation mechanisms to secure the growing ecosystem of distributed energy resources against such sophisticated, stealthy threats.