Nearly every aspect of modern society relies on large and complex power grids to deliver energy. Major disruption to these grids could have significant effects, including loss of human life and economic instability. Climate change is exacerbating the situation by making extreme weather events more powerful and frequent.
“Natural disasters and interference from bad actors on power grid infrastructure are becoming more and more prevalent,” says Russell Bent, a scientist at Los Alamos National Laboratory who uses artificial intelligence and operations research to optimize infrastructure networks. “There are constantly news articles about threats like hurricanes, wildfires, and earthquakes impacting the power grid.”
One way to improve grid resiliency is by installing microgrids. Microgrids are localized energy grids that contain their own power-generating systems and load storage capabilities. They can operate independently from the main power grid, reducing dependence on central power systems and decreasing the effects of localized, outage-inducing extreme events. The United States currently has 4 gigawatts of microgrid capacity, and 20 percent of that capacity has been added in the past five years.
“But the computational tools to analyze the impacts and potential advanced usage of these microgrids are still lagging,” says Los Alamos scientist David Fobes.
Fobes and Bent are part of the development team for PowerModelsONM, a modeling software that allows users to analyze the design, validation, and installation of grid systems. The software is a collaboration among Los Alamos, the National Renewable Energy Laboratory, Sandia National Laboratories, and the National Rural Electric Co-op Association. It was funded by the Microgrid Research and Development Program in the Department of Energy’s Office of Electricity and recently won a 2023 R&D 100 award.
“Previous technology has largely focused on the expansion-planning aspect or decision support around adding microgrids under typical operating conditions,” Fobes says. “In contrast, PowerModelsONM considers the operational strategies themselves to produce the best actions to take during extreme events or even to improve grid operations under normal conditions.”
One of the key capabilities of PowerModelsONM is the ability to model networked microgrids. Microgrids can be networked together so that excess capacity on one microgrid can support emergency backup and jumpstart efforts for another microgrid following an outage. Prior to PowerModelsONM, no software was capable of analyzing the benefits of networking microgrids together. By introducing this capability, users can generate quantifiable data on the financial benefits and resiliency benefits of networked microgrids.
PowerModelsONM was also designed to work with existing technologies, including Hardware-in-the-Loop, a small-scale testing system that allows researchers to simulate how grids might respond in various circumstances. The seamless integration between such platforms allows stakeholders to validate grids to ensure customers will be as safe as possible during an emergency.
“Combined with our rigorous validation and verification,” Fobes explains, “operators can confidently test and simulate restoration plans without risk to the electric grid.” ★