Mitigation via Analytics for Inverter-Grid Cybersecurity (MAGIC)

Feb 27, 2024

Background

The Mitigation via Analytics for Grid-Inverter Cybersecurity (MAGIC) project will develop secure Artificial Intelligence/Machine Learning (AI/ML) tools to both detect and mitigate cyber attacks on aggregations of Distributed Energy Resources (DER) in electric power distribution systems and microgrids. In so doing, MAGIC will facilitate detecting cyber attacks on DER in their earliest stages and ameliorating the effect of attacks immediately.

Objectives

  1. Develop secure AI/ML algorithms to detect cyber attacks on aggregations of DER and distinguish attacks from normal operating conditions.
  2. Extend a reinforcement learning framework developed in previous RMT/CEDS projects to mitigate the effect of cyber attacks on DER in a wide array of operating conditions.
  3. Integrate the attack detection and mitigation algorithms into a commercially available substation/microgrid management platform for algorithm demonstration.
  4. Create an open source simulation tool allowing electric utilities to determine rules to detect and mitigate cyber attacks designed to severely disrupt normal grid operations or cause voltage instabilities.
  5. Develop a software test harness to assess the security of AI/ML algorithms for electric grid attack detection and mitigation.

Project Description

The Supervisory Parameter Adjustment for Distribution Energy Storage (SPADES) project will develop the methodology and tools allowing Energy Storage Systems (ESS) to automatically reconfigure themselves to counteract cyberattacks against both the ESS control system directly and indirectly through the electric distribution grid. This research will begin with an effort to analyze the stability of both ESS control systems and the interaction of the ESS control system and the electric grid, to determine what parameters an attacker would change if a given device (or multiple devices) were to be compromised. Then, the research team will develop a supervisory control framework that utilizes adaptive control and reinforcement learning techniques to adjust ESS control system parameters and ESS active and reactive power injections to actively defend against a variety of cyberattacks. The supervisory control framework will be validated in Hardware-in-the-Loop (HIL) experiments where an independent red team will attempt to alter control parameters of an ESS to prevent the device from providing grid services. The reinforcement learning defensive algorithms will be integrated into the National Rural Electric Cooperative Association (NRECA) Open Modeling Framework (OMF), thereby allowing defensive strategies to be tailored on a utility specific basis. The major outcomes of this project will be the tools to isolate the component of the ESS control system that has been compromised during a cyberattack as well as policies for changing the control parameters of ESS to mitigate a wide variety of cyberattacks on both the ESS device itself and the electric distribution grid.

This project is supported by the U.S. Department of Energy’s Office of Cybersecurity, Energy Security, and Emergency Response (CESER) Risk Management Tools and Technologies (RMT) Program.

Principal Investigator:

Daniel Arnold (PI; LBNL)

Senior Personnel:

Sean Peisert (LBNL)
Ryan King (NREL)
Lisa Slaughter (NRECA)
Anna Scaglione (Cornell Tech)
Bruno Leao (Siemens)

Press regarding this project:

DOE Press Release: “DOE Announces $39 Million in Research Funding to Enhance Cybersecurity of Clean Distributed Energy Resources” - Sept. 12, 2023

Berkeley Lab leading the way with new cybersecurity projects - Nov. 6, 2023

More information is available on other Berkeley Lab R&D projects focusing on cybersecurity in general, as well as specifically on cybersecurity for energy delivery systems.

Power Grid Machine Learning