TR2019-035
Distributed Estimation and Detection of Cyber-Physical Attacks in Power Systems
-
- "Distributed Estimation and Detection of Cyber-Physical Attacks in Power Systems", IEEE International Conference on Communications Workshops (ICC), DOI: 10.1109/ICCW.2019.8756653, May 2019, pp. 1-6.BibTeX TR2019-035 PDF
- @inproceedings{Minot2019may,
- author = {Minot, Ariana and Sun, Hongbo and Nikovski, Daniel N. and Zhang, Jinyun},
- title = {Distributed Estimation and Detection of Cyber-Physical Attacks in Power Systems},
- booktitle = {IEEE International Conference on Communications Workshops (ICC)},
- year = 2019,
- pages = {1--6},
- month = may,
- doi = {10.1109/ICCW.2019.8756653},
- issn = {2474-9133},
- url = {https://www.merl.com/publications/TR2019-035}
- }
,
- "Distributed Estimation and Detection of Cyber-Physical Attacks in Power Systems", IEEE International Conference on Communications Workshops (ICC), DOI: 10.1109/ICCW.2019.8756653, May 2019, pp. 1-6.
-
MERL Contacts:
-
Research Area:
Abstract:
Dynamic state estimation, enabled by phasor measurement units (PMUs), opens new opportunities to improve detection of cyber-physical attacks in power networks. Distributed approaches to estimation and attack detection have many advantages, such as reduced processing times and increased security, and are arguably necessary for large size networks. In this work, we present a fully-distributed dynamic state estimation algorithm using PMU measurement data. The dynamic state estimation is jointly designed with an innovation-based attack detection scheme to limit communication overhead. An attractive feature of our work is that each control area utilizes a local model of reduced dimension. The design of our algorithm uses an approximation to the state covariance matrix, which allows for a trade-off between computation, communication, and accuracy. In numerical experiments, we demonstrate the effectiveness of this approach.