Location
Grand Wailea, Hawaii
Event Website
https://hicss.hawaii.edu/
Start Date
7-1-2020 12:00 AM
End Date
10-1-2020 12:00 AM
Description
Increasing reliance on cyber-components for communication and control has made cybersecurity in power systems an increasing concern. While Information technology (IT) based detection and prevention methods are deployed to detect cyber threats, leveraging of the system physics provides a complementary detection scheme. Here, we consider malicious power order threats directed at a high-voltage direct current (HVDC) line in a large AC network. A fast, approximate tracking state estimation method is investigated that uses a reduced iteration count and measurement prioritization using power transfer distribution factors (PTDF) to rapidly compute the approximate injections at the AC buses of the HVDC line as a power order is executed. The algorithm’s accuracy in tracking the system’s change is investigated. It is observed that with the above methods, the estimator can achieve results within 5% of the true injection. Deviations from the expected injection can be understood to be indicative of a compromised power order.
Fast, Approximate State Estimation Based Approach for Cyber Threat Detection in Power Systems
Grand Wailea, Hawaii
Increasing reliance on cyber-components for communication and control has made cybersecurity in power systems an increasing concern. While Information technology (IT) based detection and prevention methods are deployed to detect cyber threats, leveraging of the system physics provides a complementary detection scheme. Here, we consider malicious power order threats directed at a high-voltage direct current (HVDC) line in a large AC network. A fast, approximate tracking state estimation method is investigated that uses a reduced iteration count and measurement prioritization using power transfer distribution factors (PTDF) to rapidly compute the approximate injections at the AC buses of the HVDC line as a power order is executed. The algorithm’s accuracy in tracking the system’s change is investigated. It is observed that with the above methods, the estimator can achieve results within 5% of the true injection. Deviations from the expected injection can be understood to be indicative of a compromised power order.
https://aisel.aisnet.org/hicss-53/es/resillient_networks/4