Spatio-Temporal Deep Graph Network for Event Detection, Localization, and Classification in Cyber-Physical Electric Distribution System

Journal Article · Mon Jul 03 00:00:00 EDT 2023 · IEEE Transactions on Industrial Informatics

DOI:https://doi.org/10.1109/tii.2023.3290942· OSTI ID:2480941

Ahmed, Arman ^[1]; Basumallik, Sagnik ^[2]; Gholami, Amir ^[3]; Sadanandan, Sajan K. ^[3]; Namaki, Mohammad N. ^[3]; Srivastava, Anurag K. ^[3]; Wu, Yinghui ^[3]

---

1. Washington State University (WSU), Pullman, WA (United States); West Virginia University
2. West Virginia University (WVU), Morgantown, WV (United States)
3. Washington State University (WSU), Pullman, WA (United States)

This work proposes a deep graph learning framework to identify, locate, and classify power, cyber, and cyber power events at the distribution system level. The proposed algorithm jointly exploits spatial, temporal, and node-level cyber and physical data features. The developed graph neural network, together with a deep autoencoder, utilizes physical measurements from distribution level phasor measurement units and cyber data from communication network logs. The spatial structure of the synchrophasor measurements and network is incorporated through a weighted adjacency matrix. The temporal structure is incorporated by defining a spatial operation in the gated recurrent unit. This spatio-temporal learning element resides inside a power event detection, localization, and classification module that provides the degree of confidence for an event label. To accurately pinpoint the location of an event to the nearest bus equipped with a measurement unit, a combination of squared error and proximity score is utilized. Also included is a cyber event detection module that employs heteroskedasticity to analyze the significance of various cyber features during different types of attacks. Finally, a dual-bit cyber-power decision table determines the nature of the event. The proposed method is validated on two distribution systems modeled in OPAL-RT/Hypersim with limited phasor measurement units for different possible physical and cyber events. Further analyses include comparison with other state-of-the-art methods and validation in the presence of measurement noise. As a result, our method outperforms existing approaches and achieves an average detection accuracy of 97.97%, F1-score of 96.88%, precision of 96.53%, and recall of 98.57%.

• Research Organization: University of Utah, Salt Lake City, UT (United States)
• Sponsoring Organization: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Renewable Power Office. Solar Energy Technologies Office
• Grant/Contract Number: EE0008775; IA0000025
• OSTI ID: 2480941
• Report Number(s): DOE-UoU-WVU
• Journal Information: IEEE Transactions on Industrial Informatics, Journal Name: IEEE Transactions on Industrial Informatics Journal Issue: 2 Vol. 20; ISSN 1551-3203
• Publisher: IEEE
• Country of Publication: United States
• Language: English

Related Subjects

24 POWER TRANSMISSION AND DISTRIBUTION
Autoencoder
cyber-physical attack
deep learning
distribution phasor measurement unit
graph neural network