Smart grid infrastructure represents one of the most critical applications of digital twin technology for cybersecurity, as power grid disruptions can have cascading effects across all sectors of society. This paper presents a real-time cybersecurity monitoring system that leverages digital twin technology specifically designed for smart grid environments. The system maintains synchronized virtual models of grid components, enabling real-time comparison between expected and observed system states for anomaly detection. We introduce a grid-specific threat model that accounts for the unique characteristics of power system cyberattacks, including load redistribution attacks, relay manipulation, and measurement injection. The monitoring system implements a multi-granularity detection approach that operates at component, substation, and system levels simultaneously. Experimental evaluation on a realistic smart grid testbed demonstrates detection accuracy of 95.7% with average detection latency of 1.8 seconds for critical attack scenarios.
Safa Mohamed (Fri,) studied this question.