University of Georgia (UGA) researchers unveiled a sensor system that watches power electronic converters at solar energy farms for signs of cyberattack in real time.
The system can detect anomalies in a converter's operations using just one voltage sensor and one current sensor, applying deep learning methods to differentiate between normal conditions, open-circuit faults, short-circuit faults, and cyberattacks.
A passive sensor linked to the power converter gathers data on electrical waveforms and feeds it to a computer monitor, and unusual activity is detectable in the converter's electrical current, even if the firewall or security software misses an attack.
The system also can diagnose the nature of a problem, and the researchers said it can identify cyberattacks in a solar farm model more proficiently than current techniques.
From UGA Today
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA