Protecting the Grid with Artificial Intelligence

Strategic Insights -

Researchers at Sandia National Laboratories have developed brain-inspired AI algorithms that detect physical problems, cyberattacks, and both at the same time within the grid. The neural-network AI can run on inexpensive single-board computers or existing smart grid devices, providing a cost-efficient solution to protect the grid. The team collaborated with experts at Texas A&M University to create secure communication methods, particularly between grids owned by different companies. The AI can monitor the grid for abnormalities and provide early alerts of cyberattacks or physical issues.

Key Takeaways:

  • The neural-network AI can detect physical problems, cyberattacks, and both at the same time within the grid, providing a comprehensive solution to protect the grid.
  • The AI can run on inexpensive single-board computers or existing smart grid devices, making it a cost-efficient solution.
  • The team collaborated with experts at Texas A&M University to create secure communication methods, particularly between grids owned by different companies.
  • The AI can monitor the grid for abnormalities and provide early alerts of cyberattacks or physical issues.
  • The team used an autoencoder neural network, which classifies the combined data to determine whether it fits with the pattern of normal behavior or if there are abnormalities with the cyber data, physical data, or both.
  • The neural network can detect a variety of cyberattacks or physical disruptions, including denial-of-service attacks and false-data-injection attacks.
  • The AI can be tested in different environments, including emulation and hardware-in-the-loop testing, to evaluate its performance in real-world scenarios.
  • The team is working with Sierra Nevada Corporation to test how Sandia's autoencoder AI works on the company's existing cybersecurity device called Binary Armor.

Statistics:

  • The neural network can process data at a rate of 60 times per second, allowing it to detect physical problems and cyberattacks in real-time.
  • The AI can detect abnormalities in cyber data, physical data, and both, providing a comprehensive solution to protect the grid.
  • The autoencoder neural network can classify the combined data to determine whether it fits with the pattern of normal behavior or if there are abnormalities.
  • The neural network can detect a variety of cyberattacks or physical disruptions, including denial-of-service attacks and false-data-injection attacks.
  • The team is working on expanding the autoencoder AI to protect other critical infrastructure systems, such as water and natural gas distribution systems, factories, and data centers.

Sources:

  • Sandia National Laboratories: "Protecting the Grid with Artificial Intelligence" (original text)
  • Texas A&M University: (mentioned in the original text)
  • Sierra Nevada Corporation: (mentioned in the original text)
  • "R&D 100 Award-winning project called the Proactive Intrusion Detection and Mitigation System" (mentioned in the original text)