TR2020-058

Artificial Intelligence-Based Distributed Belief Propagation and Recurrent Neural Network Algorithm for Wide-Area Monitoring Systems


    •  Bhamidipati, S., Kim, K.J., Sun, H., Orlik, P.V., "Artificial Intelligence-Based Distributed Belief Propagation and Recurrent Neural Network Algorithm for Wide-Area Monitoring Systems", IEEE Network, May 2020.
      BibTeX TR2020-058 PDF
      • @article{Bhamidipati2020may,
      • author = {Bhamidipati, Sriramya and Kim, Kyeong Jin and Sun, Hongbo and Orlik, Philip V.},
      • title = {Artificial Intelligence-Based Distributed Belief Propagation and Recurrent Neural Network Algorithm for Wide-Area Monitoring Systems},
      • journal = {IEEE Network},
      • year = 2020,
      • month = may,
      • url = {https://www.merl.com/publications/TR2020-058}
      • }
  • MERL Contacts:
  • Research Areas:

    Artificial Intelligence, Communications, Multi-Physical Modeling, Signal Processing

To monitor the power grid over a wide-area, the wide-area monitoring system (WAMS) has been developed. At each substation, the Global Positioning System (GPS) receiving system resides to provide a trusted timing. Thus, it is critical for the WAMS to maintain an authentic GPS timing over a widearea. However, the GPS timing is susceptible to spoofing due to the unencrypted signal structure and its low signal power. Thus, to obtain the trusted GPS timing from spoofing, a new wide-area monitoring algorithm, which is comprised of distributed belief propagation (BP) and a bi-directional recurrent neural network (RNN), is developed under the frame of Artificial Intelligence (AI). This joint BP-RNN algorithm authenticates each power substation by evaluating the estimated GPS timing error by its distributed processing capability. Especially, the bi-directional RNN provides a fast timing error estimation under the frame of AI. Simulation results validate the fast detection time over the KullbackÔÇôLeibler divergencebased approach, and timing error estimation accuracy over the limit provided by the IEEE C37.118.1-2011 standard.