Phasor measurement units (PMUs) have been deployed at over 2,500 locations across the nation’s bulk power systems. Due to these PMU measurements, grid owners and operators now possess unprecedented quantities of data detailing the condition of the grid. Robust tools are needed for the analysis and discovery of the information hiding within the growing archives of PMU data, so future PMUs can contribute even more to the efficient, safe, and reliable operation and design of the nation’s electric system.
In 2019, the Department of Energy (DOE) selected eight projects to explore the use of big data, artificial intelligence (AI), and machine-learning technology and tools on PMU data to identify and improve existing knowledge, and to discover new insights and tools for better grid operation and management.
The projects received a first-of-its-kind dataset that:
- Covered a significant number of PMUs and substations in each of the three U.S. interconnections;
- Covered multiple years and included event logs;
- Consisted of real field data from a variety of sources that includes errors, inconsistencies, quality levels, and flaws; and
- Was anonymized to reduce the possibility of exposing potential information about power system operational vulnerabilities.
The projects are coming to completion with their final reports below along with links to some of the publications supported through this research and related presentations. After the completion of the projects, a meta report will be uploaded to this page that summarizes the key findings across the projects.
Table of Awardees
|Prime Recipient||Team Members||Project Title|
|Iowa State University of Science & Technology||Electric Power Group, Google Brain, IBM||Robust Learning of Dynamic Interactions for Enhancing Power System Resilience|
|Schweitzer Engineering Laboratories, Inc.||Oregon State University||Machine Learning Guided Operational Intelligence from Synchrophasors|
|The Regents of the University of California||Electric Power Group, Michigan Technological University||Discovery of Signatures, Anomalies, and Precursors in Synchrophasor Data with Matrix Profile and Deep Recurrent Neural Networks|
|Board of Regents, NSHE obo University of Nevada, Reno||Arizona State University, IBM, Virginia Tech||A Robust Event Diagnostics Platform: Integrating Tensor Analytics and Machine Learning Into Real-time Grid Monitoring|
|General Electric Company||GE Grid Solutions||PMU-Based Data Analytics using Digital Twin and PhasorAnalytics Software|
|Siemens Corporation, Corporate Technology||Siemens Digital Grid, Siemens Industries and Drives (Mindsphere), Southern Methodist University, Temple University||MindSynchro|
|Ping Things, Inc.||N/A||Combinatorial Evaluation of Physical Feature Engineering and Deep Temporal Modeling for Synchrophasor Data at Scale|
|Texas A&M Engineering Experiment Station||Temple University, Quanta Technology LLC, OSIsoft, LLC||Big Data Synchrophasor Monitoring and Analytics for Resiliency Tracking (BDSMART)|
- GE Final Scientific and Technical Report
- More to come.
Related Publications and Presentations
- Ping Things Presentation at PES re FOA 1861 ProjectPingThings, Inc. Panel talk Discussing the Big Data FOA 1861 Project
- Siemens Presentation at IEEE Big Data Conference
- Siemens Big Data Processing for Power Grid Event Detection
- FOA 1861 Awardees report out on Machine Learning and AI GE, SEL, Siemens, UNRFOA 1861 Awardees Report on Machine Learning and AI
- PingThings, Inc. FOA 1861 Final Presentation
- Data for FOA 1861 - Bringing It All Together
If you have any questions regarding the data used for these projects, please contact Jim Follum at firstname.lastname@example.org.