Lawrence Livermore National Laboratory (LLNL) will receive funding from the Department of Energy (DOE) over the next three years to strengthen, transform and improve the nation’s energy infrastructure.
LLNL is the lead on three projects:
Using sensors to identify faulty or aging components of the nation’s energy grid system
In its first stage, this application will identify urgent failures, dealing with the local sensing and “on the ground” analytics, and improve detection and maintenance plans for growing and emerging failures, addressing their root cause and providing active sensing diagnostics and predictions when the systems might fail in the future.
By identifying maintenance issues that will lead to failure, from multi-modal sources of data including new and existing sensor platforms, aerial photography, infrared scans and chemical sensing, the application can reduce the risk of power lines igniting a fire under certain conditions. The sensing and analytics application would allow power to stay on and only turn off lines that pose an actual risk.
“On this project, we are using a hybridization of different machine learning algorithms to simultaneously fuse information from different sensors and learn the underlying temporal structure of the data,” said Brenda Ng, LLNL group leader for Machine Learning. “Our mathematical framework of temporal multimodal learning will be used to infer the underlying health state of the energy grid component. The framework has its origin as an LLNL-developed project and has been applied to patient health prediction from electronic healthcare record data.”
National lab partners include Oak Ridge National Laboratory, Sandia National Laboratories and National Energy Technology Laboratory.
Using healthcare analytics for critical grid component diagnostics
This project will leverage deep reinforcement learning to understand the “healthy” state of the system and detect bad behaviors and events in the grid that would disrupt that state. By collecting network and power system data, the algorithms will be trained to recognize and attribute bad behaviors resulting from cyber attacks ensuring appropriate and timely response.
“We’re using artificial intelligence to create an immune system (much like the body’s immune system that identifies and fights off pathogens) that would automatically identify when there is something faulty in the grid and identify the root cause of the failure, specifically focusing on cyber attacks. It would then come up with a series of responses to help it fix or safely isolate parts of the system that have been compromised,” said Jovana Helms, LLNL associate program leader for Civilian Cybersecurity. ”The really cool thing about this approach is that the algorithms continue to be trained after they are deployed, so the system continues to learn and gets better over time making our grid truly ‘intelligent.’”
National lab partners include Idaho National Laboratory, Oak Ridge National Laboratory and Sandia National Laboratories.
Energy storage and system flexibility
A large-scale black out, with restoration hampered by uncertainties, cyber attacks and communication breakdown would disrupt the nation’s security for a prolonged period.
This project builds upon the DARPA RADICS program, to exercise the capabilities of distributed resources in a cyber restoration at the Plum Island facility and develop procedural documentation for future implementation across the nation. Plum Island is a U.S. federal research facility that is part of the Department of Homeland Security's Directorate for Science and Technology.
“We’ll look at how we use the distributed energy to make our grid more resilient after any event,” said Emma Stewart, LLNL associate program leader for Defense Infrastructure. “It can take a long time to get the power back up when there are significant interconnected infrastructure components. We could leverage a distributed grid to start to bring the power back up from the outside in rather than a central location.”
Nation lab partners include Pacific Northwest National Laboratory and Idaho National Laboratory.
The projects are part of DOE’s Grid Modernization Initiative (GMI), a crosscutting effort that focuses public and private partnerships to develop a portfolio of new tools and technologies that measure, analyze, predict, protect and control the grid of the future. This solicitation is focused on developing projects in resilience modeling, energy storage and system flexibility, advanced sensors and data analytics, institutional support and analysis, cyber-physical security and generation.
“In order to continue growing our economy and strengthening our national security, we need to invest in projects that bolster the reliability and resiliency of our grid,” said Under Secretary of Energy Mark Menezes. “Working together with our national labs and private sector partners, these projects will establish a stronger, more resilient grid and ensure that Americans continue to enjoy the benefits of unprecedented energy abundance.”
LLNL is participating in a total of nine of the 23 projects funded.
At least two of the DOE’s five applied science offices (the Office of Cybersecurity, Energy Security and Emergency Response; the Office of Electricity; the Office of Energy Efficiency and Renewable Energy; the Office of Fossil Energy; and the Office of Nuclear Energy) will sponsor each project.
The projects must have impact on the bulk power system and have near-term success to increase likelihood of industry implementation and increase potential impact. The projects will provide meaningful results within 18 months to two years that address clear, immediate challenges.
The Worldwide Threat Assessment of the U.S. Intelligence Community report — presented to Congress by the Director of National Intelligence in January 2019 — identified threats to U.S. national security which will expand and diversify in the coming years. The development and application of new technologies will introduce both risks and opportunities across the U.S. economy, and the Department of Homeland Security has identified the energy sector as one of 16 critical infrastructure sectors in the nation. The U.S. electric power system must prepare for and adapt to changing conditions and withstand and recover rapidly from disruptions. GMI has focused on security challenges of the electric power system in their upcoming portfolio of projects.
The Grid Modernization Laboratory Consortium (GMLC) was established as a strategic partnership between DOE and the national laboratories to bring together leading experts, technologies and resources to collaborate on the goal of modernizing the nation’s grid. The benefits of the GMLC include more efficient use of resources, shared networks, better learning and preservation of knowledge, enhanced lab coordination and collaboration and regional perspective and relationships with local stakeholders and industry.
stark8 [at] llnl.gov