Abstract:
The integration of Internet of Things (IoT) devices in electric power information systems
has introduced unprecedented cybersecurity challenges. This study develops and evaluates a
comprehensive cybersecurity framework tailored for IoT-integrated power grids, addressing
the unique vulnerabilities and complexities of these critical systems. A multi-layered security approach was designed, incorporating device authentication, encrypted communication,
and machine learning-based anomaly detection. The framework underwent extensive testing
across six distinct attack types (unauthorized access, man-in-the-middle, DDoS, malicious
command injection, firmware tampering, and data exfiltration), with over 10,000 simulated
attack scenarios conducted in a testbed environment mimicking a regional power grid with
up to 10,000 IoT devices. The framework demonstrated high effectiveness, with average
threat detection rates of 97.9% and prevention rates of 97.1% across all attack vectors. Performance testing revealed sub-linear CPU utilization growth as IoT devices scaled from 100
to 10,000, with only a 2.3% increase in network latency at the 1,000-device scale. The system
maintained 98.7-99.8% availability during attacks and achieved 94-98% compliance with key
industry standards. These findings demonstrate the framework's robust capabilities in securing IoT-integrated power systems while highlighting areas for future research in extreme
scalability scenarios and real-world implementation challenges.
