Lincoln Laboratory

Subterms

    Latest story

    50 Shares129 Views

    Confronting the AI/energy conundrum

    by

    The explosive growth of AI-powered computing centers is creating an unprecedented surge in electricity demand that threatens to overwhelm power grids and derail climate goals. At the same time, artificial intelligence technologies could revolutionize energy systems, accelerating the transition to clean power.“We’re at a cusp of potentially gigantic change throughout the economy,” said William H. Green, director of the MIT Energy Initiative (MITEI) and Hoyt C. Hottel Professor in the MIT Department of Chemical Engineering, at MITEI’s Spring Symposium, “AI and energy: Peril and promise,” held on May 13. The event brought together experts from industry, academia, and government to explore solutions to what Green described as both “local problems with electric supply and meeting our clean energy targets” while seeking to “reap the benefits of AI without some of the harms.” The challenge of data center energy demand and potential benefits of AI to the energy transition is a research priority for MITEI.AI’s startling energy demandsFrom the start, the symposium highlighted sobering statistics about AI’s appetite for electricity. After decades of flat electricity demand in the United States, computing centers now consume approximately 4 percent of the nation’s electricity. Although there is great uncertainty, some projections suggest this demand could rise to 12-15 percent by 2030, largely driven by artificial intelligence applications.Vijay Gadepally, senior scientist at MIT’s Lincoln Laboratory, emphasized the scale of AI’s consumption. “The power required for sustaining some of these large models is doubling almost every three months,” he noted. “A single ChatGPT conversation uses as much electricity as charging your phone, and generating an image consumes about a bottle of water for cooling.”Facilities requiring 50 to 100 megawatts of power are emerging rapidly across the United States and globally, driven both by casual and institutional research needs relying on large language programs such as ChatGPT and Gemini. Gadepally cited congressional testimony by Sam Altman, CEO of OpenAI, highlighting how fundamental this relationship has become: “The cost of intelligence, the cost of AI, will converge to the cost of energy.”“The energy demands of AI are a significant challenge, but we also have an opportunity to harness these vast computational capabilities to contribute to climate change solutions,” said Evelyn Wang, MIT vice president for energy and climate and the former director at the Advanced Research Projects Agency-Energy (ARPA-E) at the U.S. Department of Energy.Wang also noted that innovations developed for AI and data centers — such as efficiency, cooling technologies, and clean-power solutions — could have broad applications beyond computing facilities themselves.Strategies for clean energy solutionsThe symposium explored multiple pathways to address the AI-energy challenge. Some panelists presented models suggesting that while artificial intelligence may increase emissions in the short term, its optimization capabilities could enable substantial emissions reductions after 2030 through more efficient power systems and accelerated clean technology development.Research shows regional variations in the cost of powering computing centers with clean electricity, according to Emre Gençer, co-founder and CEO of Sesame Sustainability and former MITEI principal research scientist. Gençer’s analysis revealed that the central United States offers considerably lower costs due to complementary solar and wind resources. However, achieving zero-emission power would require massive battery deployments — five to 10 times more than moderate carbon scenarios — driving costs two to three times higher.“If we want to do zero emissions with reliable power, we need technologies other than renewables and batteries, which will be too expensive,” Gençer said. He pointed to “long-duration storage technologies, small modular reactors, geothermal, or hybrid approaches” as necessary complements.Because of data center energy demand, there is renewed interest in nuclear power, noted Kathryn Biegel, manager of R&D and corporate strategy at Constellation Energy, adding that her company is restarting the reactor at the former Three Mile Island site, now called the “Crane Clean Energy Center,” to meet this demand. “The data center space has become a major, major priority for Constellation,” she said, emphasizing how their needs for both reliability and carbon-free electricity are reshaping the power industry.Can AI accelerate the energy transition?Artificial intelligence could dramatically improve power systems, according to Priya Donti, assistant professor and the Silverman Family Career Development Professor in MIT’s Department of Electrical Engineering and Computer Science and the Laboratory for Information and Decision Systems. She showcased how AI can accelerate power grid optimization by embedding physics-based constraints into neural networks, potentially solving complex power flow problems at “10 times, or even greater, speed compared to your traditional models.”AI is already reducing carbon emissions, according to examples shared by Antonia Gawel, global director of sustainability and partnerships at Google. Google Maps’ fuel-efficient routing feature has “helped to prevent more than 2.9 million metric tons of GHG [greenhouse gas] emissions reductions since launch, which is the equivalent of taking 650,000 fuel-based cars off the road for a year,” she said. Another Google research project uses artificial intelligence to help pilots avoid creating contrails, which represent about 1 percent of global warming impact.AI’s potential to speed materials discovery for power applications was highlighted by Rafael Gómez-Bombarelli, the Paul M. Cook Career Development Associate Professor in the MIT Department of Materials Science and Engineering. “AI-supervised models can be trained to go from structure to property,” he noted, enabling the development of materials crucial for both computing and efficiency.Securing growth with sustainabilityThroughout the symposium, participants grappled with balancing rapid AI deployment against environmental impacts. While AI training receives most attention, Dustin Demetriou, senior technical staff member in sustainability and data center innovation at IBM, quoted a World Economic Forum article that suggested that “80 percent of the environmental footprint is estimated to be due to inferencing.” Demetriou emphasized the need for efficiency across all artificial intelligence applications.Jevons’ paradox, where “efficiency gains tend to increase overall resource consumption rather than decrease it” is another factor to consider, cautioned Emma Strubell, the Raj Reddy Assistant Professor in the Language Technologies Institute in the School of Computer Science at Carnegie Mellon University. Strubell advocated for viewing computing center electricity as a limited resource requiring thoughtful allocation across different applications.Several presenters discussed novel approaches for integrating renewable sources with existing grid infrastructure, including potential hybrid solutions that combine clean installations with existing natural gas plants that have valuable grid connections already in place. These approaches could provide substantial clean capacity across the United States at reasonable costs while minimizing reliability impacts.Navigating the AI-energy paradoxThe symposium highlighted MIT’s central role in developing solutions to the AI-electricity challenge.Green spoke of a new MITEI program on computing centers, power, and computation that will operate alongside the comprehensive spread of MIT Climate Project research. “We’re going to try to tackle a very complicated problem all the way from the power sources through the actual algorithms that deliver value to the customers — in a way that’s going to be acceptable to all the stakeholders and really meet all the needs,” Green said.Participants in the symposium were polled about priorities for MIT’s research by Randall Field, MITEI director of research. The real-time results ranked “data center and grid integration issues” as the top priority, followed by “AI for accelerated discovery of advanced materials for energy.”In addition, attendees revealed that most view AI’s potential regarding power as a “promise,” rather than a “peril,” although a considerable portion remain uncertain about the ultimate impact. When asked about priorities in power supply for computing facilities, half of the respondents selected carbon intensity as their top concern, with reliability and cost following. More

    More stories

    • 88 Shares119 Views

      in Energy

      Q&A: The climate impact of generative AI

      by

      Vijay Gadepally, a senior staff member at MIT Lincoln Laboratory, leads a number of projects at the Lincoln Laboratory Supercomputing Center (LLSC) to make computing platforms, and the artificial intelligence systems that run on them, more efficient. Here, Gadepally discusses the increasing use of generative AI in everyday tools, its hidden environmental impact, and some of the ways that Lincoln Laboratory and the greater AI community can reduce emissions for a greener future.Q: What trends are you seeing in terms of how generative AI is being used in computing?A: Generative AI uses machine learning (ML) to create new content, like images and text, based on data that is inputted into the ML system. At the LLSC we design and build some of the largest academic computing platforms in the world, and over the past few years we’ve seen an explosion in the number of projects that need access to high-performance computing for generative AI. We’re also seeing how generative AI is changing all sorts of fields and domains — for example, ChatGPT is already influencing the classroom and the workplace faster than regulations can seem to keep up.We can imagine all sorts of uses for generative AI within the next decade or so, like powering highly capable virtual assistants, developing new drugs and materials, and even improving our understanding of basic science. We can’t predict everything that generative AI will be used for, but I can certainly say that with more and more complex algorithms, their compute, energy, and climate impact will continue to grow very quickly.Q: What strategies is the LLSC using to mitigate this climate impact?A: We’re always looking for ways to make computing more efficient, as doing so helps our data center make the most of its resources and allows our scientific colleagues to push their fields forward in as efficient a manner as possible.As one example, we’ve been reducing the amount of power our hardware consumes by making simple changes, similar to dimming or turning off lights when you leave a room. In one experiment, we reduced the energy consumption of a group of graphics processing units by 20 percent to 30 percent, with minimal impact on their performance, by enforcing a power cap. This technique also lowered the hardware operating temperatures, making the GPUs easier to cool and longer lasting.Another strategy is changing our behavior to be more climate-aware. At home, some of us might choose to use renewable energy sources or intelligent scheduling. We are using similar techniques at the LLSC — such as training AI models when temperatures are cooler, or when local grid energy demand is low.We also realized that a lot of the energy spent on computing is often wasted, like how a water leak increases your bill but without any benefits to your home. We developed some new techniques that allow us to monitor computing workloads as they are running and then terminate those that are unlikely to yield good results. Surprisingly, in a number of cases we found that the majority of computations could be terminated early without compromising the end result.Q: What’s an example of a project you’ve done that reduces the energy output of a generative AI program?A: We recently built a climate-aware computer vision tool. Computer vision is a domain that’s focused on applying AI to images; so, differentiating between cats and dogs in an image, correctly labeling objects within an image, or looking for components of interest within an image.In our tool, we included real-time carbon telemetry, which produces information about how much carbon is being emitted by our local grid as a model is running. Depending on this information, our system will automatically switch to a more energy-efficient version of the model, which typically has fewer parameters, in times of high carbon intensity, or a much higher-fidelity version of the model in times of low carbon intensity.By doing this, we saw a nearly 80 percent reduction in carbon emissions over a one- to two-day period. We recently extended this idea to other generative AI tasks such as text summarization and found the same results. Interestingly, the performance sometimes improved after using our technique!Q: What can we do as consumers of generative AI to help mitigate its climate impact?A: As consumers, we can ask our AI providers to offer greater transparency. For example, on Google Flights, I can see a variety of options that indicate a specific flight’s carbon footprint. We should be getting similar kinds of measurements from generative AI tools so that we can make a conscious decision on which product or platform to use based on our priorities.We can also make an effort to be more educated on generative AI emissions in general. Many of us are familiar with vehicle emissions, and it can help to talk about generative AI emissions in comparative terms. People may be surprised to know, for example, that one image-generation task is roughly equivalent to driving four miles in a gas car, or that it takes the same amount of energy to charge an electric car as it does to generate about 1,500 text summarizations.There are many cases where customers would be happy to make a trade-off if they knew the trade-off’s impact.Q: What do you see for the future?A: Mitigating the climate impact of generative AI is one of those problems that people all over the world are working on, and with a similar goal. We’re doing a lot of work here at Lincoln Laboratory, but its only scratching at the surface. In the long term, data centers, AI developers, and energy grids will need to work together to provide “energy audits” to uncover other unique ways that we can improve computing efficiencies. We need more partnerships and more collaboration in order to forge ahead.If you’re interested in learning more, or collaborating with Lincoln Laboratory on these efforts, please contact Vijay Gadepally.

      Play video

      Video: MIT Lincoln Laboratory More

    • 138 Shares99 Views

      in Energy

      How hard is it to prevent recurring blackouts in Puerto Rico?

      by

      Researchers at MIT’s Laboratory for Information and Decision Systems (LIDS) have shown that using decision-making software and dynamic monitoring of weather and energy use can significantly improve resiliency in the face of weather-related outages, and can also help to efficiently integrate renewable energy sources into the grid.The researchers point out that the system they suggest might have prevented or at least lessened the kind of widespread power outage that Puerto Rico experienced last week by providing analysis to guide rerouting of power through different lines and thus limit the spread of the outage.The computer platform, which the researchers describe as DyMonDS, for Dynamic Monitoring and Decision Systems, can be used to enhance the existing operating and planning practices used in the electric industry. The platform supports interactive information exchange and decision-making between the grid operators and grid-edge users — all the distributed power sources, storage systems and software that contribute to the grid. It also supports optimization of available resources and controllable grid equipment as system conditions vary. It further lends itself to implementing cooperative decision-making by different utility- and non-utility-owned electric power grid users, including portfolios of mixed resources, users, and storage. Operating and planning the interactions of the end-to-end high-voltage transmission grid with local distribution grids and microgrids represents another major potential use of this platform.This general approach was illustrated using a set of publicly-available data on both meteorology and details of electricity production and distribution in Puerto Rico. An extended AC Optimal Power Flow software developed by SmartGridz Inc. is used for system-level optimization of controllable equipment. This provides real-time guidance for deciding how much power, and through which transmission lines, should be channeled by adjusting plant dispatch and voltage-related set points, and in extreme cases, where to reduce or cut power in order to maintain physically-implementable service for as many customers as possible. The team found that the use of such a system can help to ensure that the greatest number of critical services maintain power even during a hurricane, and at the same time can lead to a substantial decrease in the need for construction of new power plants thanks to more efficient use of existing resources.The findings are described in a paper in the journal Foundations and Trends in Electric Energy Systems, by MIT LIDS researchers Marija Ilic and Laurentiu Anton, along with recent alumna Ramapathi Jaddivada.“Using this software,” Ilic says, they show that “even during bad weather, if you predict equipment failures, and by using that information exchange, you can localize the effect of equipment failures and still serve a lot of customers, 50 percent of customers, when otherwise things would black out.”Anton says that “the way many grids today are operated is sub-optimal.” As a result, “we showed how much better they could do even under normal conditions, without any failures, by utilizing this software.” The savings resulting from this optimization, under everyday conditions, could be in the tens of percents, they say.The way utility systems plan currently, Ilic says, “usually the standard is that they have to build enough capacity and operate in real time so that if one large piece of equipment fails, like a large generator or transmission line, you still serve customers in an uninterrupted way. That’s what’s called N-minus-1.” Under this policy, if one major component of the system fails, they should be able to maintain service for at least 30 minutes. That system allows utilities to plan for how much reserve generating capacity they need to have on hand. That’s expensive, Ilic points out, because it means maintaining this reserve capacity all the time, even under normal operating conditions when it’s not needed.In addition, “right now there are no criteria for what I call N-minus-K,” she says. If bad weather causes five pieces of equipment to fail at once, “there is no software to help utilities decide what to schedule” in terms of keeping the most customers, and the most important services such as hospitals and emergency services, provided with power. They showed that even with 50 percent of the infrastructure out of commission, it would still be possible to keep power flowing to a large proportion of customers.Their work on analyzing the power situation in Puerto Rico started after the island had been devastated by hurricanes Irma and Maria. Most of the electric generation capacity is in the south, yet the largest loads are in San Juan, in the north, and Mayaguez in the west. When transmission lines get knocked down, a lot of rerouting of power needs to happen quickly.With the new systems, “the software finds the optimal adjustments for set points,” for example, changing voltages can allow for power to be redirected through less-congested lines, or can be increased to lessen power losses, Anton says.The software also helps in the long-term planning for the grid. As many fossil-fuel power plants are scheduled to be decommissioned soon in Puerto Rico, as they are in many other places, planning for how to replace that power without having to resort to greenhouse gas-emitting sources is a key to achieving carbon-reduction goals. And by analyzing usage patterns, the software can guide the placement of new renewable power sources where they can most efficiently provide power where and when it’s needed.As plants are retired or as components are affected by weather, “We wanted to ensure the dispatchability of power when the load changes,” Anton says, “but also when crucial components are lost, to ensure the robustness at each step of the retirement schedule.”One thing they found was that “if you look at how much generating capacity exists, it’s more than the peak load, even after you retire a few fossil plants,” Ilic says. “But it’s hard to deliver.” Strategic planning of new distribution lines could make a big difference.Jaddivada, director of innovation at SmartGridz, says that “we evaluated different possible architectures in Puerto Rico, and we showed the ability of this software to ensure uninterrupted electricity service. This is the most important challenge utilities have today. They have to go through a computationally tedious process to make sure the grid functions for any possible outage in the system. And that can be done in a much more efficient way through the software that the company  developed.”The project was a collaborative effort between the MIT LIDS researchers and others at MIT Lincoln Laboratory, the Pacific Northwest National Laboratory, with overall help of SmartGridz software.  More

    • 150 Shares109 Views

      in Environment

      Surface-based sonar system could rapidly map the ocean floor at high resolution

      by

      On June 18, 2023, the Titan submersible was about an hour-and-a-half into its two-hour descent to the Titanic wreckage at the bottom of the Atlantic Ocean when it lost contact with its support ship. This cease in communication set off a frantic search for the tourist submersible and five passengers onboard, located about two miles below the ocean’s surface.Deep-ocean search and recovery is one of the many missions of military services like the U.S. Coast Guard Office of Search and Rescue and the U.S. Navy Supervisor of Salvage and Diving. For this mission, the longest delays come from transporting search-and-rescue equipment via ship to the area of interest and comprehensively surveying that area. A search operation on the scale of that for Titan — which was conducted 420 nautical miles from the nearest port and covered 13,000 square kilometers, an area roughly twice the size of Connecticut — could take weeks to complete. The search area for Titan is considered relatively small, focused on the immediate vicinity of the Titanic. When the area is less known, operations could take months. (A remotely operated underwater vehicle deployed by a Canadian vessel ended up finding the debris field of Titan on the seafloor, four days after the submersible had gone missing.)A research team from MIT Lincoln Laboratory and the MIT Department of Mechanical Engineering’s Ocean Science and Engineering lab is developing a surface-based sonar system that could accelerate the timeline for small- and large-scale search operations to days. Called the Autonomous Sparse-Aperture Multibeam Echo Sounder, the system scans at surface-ship rates while providing sufficient resolution to find objects and features in the deep ocean, without the time and expense of deploying underwater vehicles. The echo sounder — which features a large sonar array using a small set of autonomous surface vehicles (ASVs) that can be deployed via aircraft into the ocean — holds the potential to map the seabed at 50 times the coverage rate of an underwater vehicle and 100 times the resolution of a surface vessel.

      Play video

      Autonomous Sparse-Aperture Multibeam Echo SounderVideo: MIT Lincoln Laboratory

      “Our array provides the best of both worlds: the high resolution of underwater vehicles and the high coverage rate of surface ships,” says co–principal investigator Andrew March, assistant leader of the laboratory’s Advanced Undersea Systems and Technology Group. “Though large surface-based sonar systems at low frequency have the potential to determine the materials and profiles of the seabed, they typically do so at the expense of resolution, particularly with increasing ocean depth. Our array can likely determine this information, too, but at significantly enhanced resolution in the deep ocean.”Underwater unknownOceans cover 71 percent of Earth’s surface, yet more than 80 percent of this underwater realm remains undiscovered and unexplored. Humans know more about the surface of other planets and the moon than the bottom of our oceans. High-resolution seabed maps would not only be useful to find missing objects like ships or aircraft, but also to support a host of other scientific applications: understanding Earth’s geology, improving forecasting of ocean currents and corresponding weather and climate impacts, uncovering archaeological sites, monitoring marine ecosystems and habitats, and identifying locations containing natural resources such as mineral and oil deposits.Scientists and governments worldwide recognize the importance of creating a high-resolution global map of the seafloor; the problem is that no existing technology can achieve meter-scale resolution from the ocean surface. The average depth of our oceans is approximately 3,700 meters. However, today’s technologies capable of finding human-made objects on the seabed or identifying person-sized natural features — these technologies include sonar, lidar, cameras, and gravitational field mapping — have a maximum range of less than 1,000 meters through water.Ships with large sonar arrays mounted on their hull map the deep ocean by emitting low-frequency sound waves that bounce off the seafloor and return as echoes to the surface. Operation at low frequencies is necessary because water readily absorbs high-frequency sound waves, especially with increasing depth; however, such operation yields low-resolution images, with each image pixel representing a football field in size. Resolution is also restricted because sonar arrays installed on large mapping ships are already using all of the available hull space, thereby capping the sonar beam’s aperture size. By contrast, sonars on autonomous underwater vehicles (AUVs) that operate at higher frequencies within a few hundred meters of the seafloor generate maps with each pixel representing one square meter or less, resulting in 10,000 times more pixels in that same football field–sized area. However, this higher resolution comes with trade-offs: AUVs are time-consuming and expensive to deploy in the deep ocean, limiting the amount of seafloor that can be mapped; they have a maximum range of about 1,000 meters before their high-frequency sound gets absorbed; and they move at slow speeds to conserve power. The area-coverage rate of AUVs performing high-resolution mapping is about 8 square kilometers per hour; surface vessels map the deep ocean at more than 50 times that rate.A solution surfacesThe Autonomous Sparse-Aperture Multibeam Echo Sounder could offer a cost-effective approach to high-resolution, rapid mapping of the deep seafloor from the ocean’s surface. A collaborative fleet of about 20 ASVs, each hosting a small sonar array, effectively forms a single sonar array 100 times the size of a large sonar array installed on a ship. The large aperture achieved by the array (hundreds of meters) produces a narrow beam, which enables sound to be precisely steered to generate high-resolution maps at low frequency. Because very few sonars are installed relative to the array’s overall size (i.e., a sparse aperture), the cost is tractable.However, this collaborative and sparse setup introduces some operational challenges. First, for coherent 3D imaging, the relative position of each ASV’s sonar subarray must be accurately tracked through dynamic ocean-induced motions. Second, because sonar elements are not placed directly next to each other without any gaps, the array suffers from a lower signal-to-noise ratio and is less able to reject noise coming from unintended or undesired directions. To mitigate these challenges, the team has been developing a low-cost precision-relative navigation system and leveraging acoustic signal processing tools and new ocean-field estimation algorithms. The MIT campus collaborators are developing algorithms for data processing and image formation, especially to estimate depth-integrated water-column parameters. These enabling technologies will help account for complex ocean physics, spanning physical properties like temperature, dynamic processes like currents and waves, and acoustic propagation factors like sound speed.Processing for all required control and calculations could be completed either remotely or onboard the ASVs. For example, ASVs deployed from a ship or flying boat could be controlled and guided remotely from land via a satellite link or from a nearby support ship (with direct communications or a satellite link), and left to map the seabed for weeks or months at a time until maintenance is needed. Sonar-return health checks and coarse seabed mapping would be conducted on board, while full, high-resolution reconstruction of the seabed would require a supercomputing infrastructure on land or on a support ship.”Deploying vehicles in an area and letting them map for extended periods of time without the need for a ship to return home to replenish supplies and rotate crews would significantly simplify logistics and operating costs,” says co–principal investigator Paul Ryu, a researcher in the Advanced Undersea Systems and Technology Group.Since beginning their research in 2018, the team has turned their concept into a prototype. Initially, the scientists built a scale model of a sparse-aperture sonar array and tested it in a water tank at the laboratory’s Autonomous Systems Development Facility. Then, they prototyped an ASV-sized sonar subarray and demonstrated its functionality in Gloucester, Massachusetts. In follow-on sea tests in Boston Harbor, they deployed an 8-meter array containing multiple subarrays equivalent to 25 ASVs locked together; with this array, they generated 3D reconstructions of the seafloor and a shipwreck. Most recently, the team fabricated, in collaboration with Woods Hole Oceanographic Institution, a first-generation, 12-foot-long, all-electric ASV prototype carrying a sonar array underneath. With this prototype, they conducted preliminary relative navigation testing in Woods Hole, Massachusetts and Newport, Rhode Island. Their full deep-ocean concept calls for approximately 20 such ASVs of a similar size, likely powered by wave or solar energy.This work was funded through Lincoln Laboratory’s internally administered R&D portfolio on autonomous systems. The team is now seeking external sponsorship to continue development of their ocean floor–mapping technology, which was recognized with a 2024 R&D 100 Award.  More

    • 113 Shares179 Views

      in Environment

      Researchers return to Arctic to test integrated sensor nodes

      by

      Shimmering ice extends in all directions as far as the eye can see. Air temperatures plunge to minus 40 degrees Fahrenheit and colder with wind chills. Ocean currents drag large swaths of ice floating at sea. Polar bears, narwhals, and other iconic Arctic species roam wild.For a week this past spring, MIT Lincoln Laboratory researchers Ben Evans and Dave Whelihan called this place — drifting some 200 nautical miles offshore from Prudhoe Bay, Alaska, on the frozen Beaufort Sea in the Arctic Circle — home. Two ice runways for small aircraft provided their only way in and out of this remote wilderness; heated tents provided their only shelter from the bitter cold.

      Play video

      Video: MIT Lincoln Laboratory

      Here, in the northernmost region on Earth, Evans and Whelihan joined other groups conducting fieldwork in the Arctic as part of Operation Ice Camp (OIC) 2024, an operational exercise run by the U.S. Navy’s Arctic Submarine Laboratory (ASL). Riding on snowmobiles and helicopters, the duo deployed a small set of integrated sensor nodes that measure everything from atmospheric conditions to ice properties to the structure of water deep below the surface.Ultimately, they envision deploying an unattended network of these low-cost sensor nodes across the Arctic to increase scientific understanding of the trending loss in sea ice extent and thickness. Warming much faster than the rest of the world, the Arctic is a ground zero for climate change, with cascading impacts across the planet that include rising sea levels and extreme weather. Openings in the sea ice cover, or leads, are concerning not only for climate change but also for global geopolitical competition over transit routes and natural resources. A synoptic view of the physical processes happening above, at, and below sea ice is key to determining why the ice is diminishing. In turn, this knowledge can help predict when and where fractures will occur, to inform planning and decision-making.Winter “camp”Every two years, OIC, previously called Ice Exercise (ICEX), provides a way for the international community to access the Arctic for operational readiness exercises and scientific research, with the focus switching back and forth; this year’s focus was scientific research. Coordination, planning, and execution of the month-long operation is led by ASL, a division of the U.S. Navy’s Undersea Warfighting Development Center responsible for ensuring the submarine force can effectively operate in the Arctic Ocean.Making this inhospitable and unforgiving environment safe for participants takes considerable effort. The critical first step is determining where to set up camp. In the weeks before the first participants arrived for OIC 2024, ASL — with assistance from the U.S. National Ice Center, University of Alaska Fairbanks Geophysical Institute, and UIC Science — flew over large sheets of floating ice (ice floes) identified via satellite imagery, landed on some they thought might be viable sites, and drilled through the ice to check its thickness. The ice floe must not only be large enough to accommodate construction of a camp and two runways but also feature both multiyear ice and first-year ice. Multiyear ice is thick and strong but rough, making it ideal for camp setup, while the smooth but thinner first-year ice is better suited for building runways. Once the appropriate ice floe was selected, ASL began to haul in equipment and food, build infrastructure like lodging and a command center, and fly in a small group before fully operationalizing the site. They also identified locations near the camp for two Navy submarines to surface through the ice.The more than 200 participants represented U.S. and allied forces and scientists from research organizations and universities. Distinguished visitors from government offices also attended OIC to see the unique Arctic environment and unfolding challenges firsthand.“Our ASL hosts do incredible work to build this camp from scratch and keep us alive,” Evans says.Evans and Whelihan, part of the laboratory’s Advanced Undersea Systems and Technology Group, first trekked to the Arctic in March 2022 for ICEX 2022. (The laboratory in general has been participating since 2016 in these events, the first iteration of which occurred in 1946.) There, they deployed a suite of commercial off-the-shelf sensors for detecting acoustic (sound) and seismic (vibration) events created by ice fractures or collisions, and for measuring salinity, temperature, and pressure in the water below the ice. They also deployed a prototype fiber-based temperature sensor array developed by the laboratory and research partners for precisely measuring temperature across the entire water column at one location, and a University of New Hampshire (UNH)−supplied echosounder to investigate the different layers present in the water column. In this maiden voyage, their goals were to assess how these sensors fared in the harsh Arctic conditions and to collect a dataset from which characteristic signatures of ice-fracturing events could begin to be identified. These events would be correlated with weather and water conditions to eventually offer a predictive capability.“We saw real phenomenology in our data,” Whelihan says. “But, we’re not ice experts. What we’re good at here at the laboratory is making and deploying sensors. That’s our place in the world of climate science: to be a data provider. In fact, we hope to open source all of our data this year so that ice scientists can access and analyze them and then we can make enhanced sensors and collect more data.”Interim iceIn the two years since that expedition, they and their colleagues have been modifying their sensor designs and deployment strategies. As Evans and Whelihan learned at ICEX 2022, to be resilient in the Arctic, a sensor must not only be kept warm and dry during deployment but also be deployed in a way to prevent breaking. Moreover, sufficient power and data links are needed to collect and access sensor data.“We can make cold-weather electronics, no problem,” Whelihan says. “The two drivers are operating the sensors in an energy-starved environment — the colder it is, the worse batteries perform — and keeping them from getting destroyed when ice floes crash together as leads in the ice open up.”Their work in the interim to OIC 2024 involved integrating the individual sensors into hardened sensor nodes and practicing deploying these nodes in easier-to-access locations. To facilitate incorporating additional sensors into a node, Whelihan spearheaded the development of an open-source, easily extensible hardware and software architecture.In March 2023, the Lincoln Laboratory team deployed three sensor nodes for a week on Huron Bay off Lake Superior through Michigan Tech’s Great Lakes Research Center (GLRC). Engineers from GLRC helped the team safely set up an operations base on the ice. They demonstrated that the sensor integration worked, and the sensor nodes proved capable of surviving for at least a week in relatively harsh conditions. The researchers recorded seismic activity on all three nodes, corresponding to some ice breaking further up the bay.“Proving our sensor node in an Arctic surrogate environment provided a stepping stone for testing in the real Arctic,” Evans says.Evans then received an invitation from Ignatius Rigor, the coordinator of the International Arctic Buoy Program (IABP), to join him on an upcoming trip to Utqiaġvik (formerly Barrow), Alaska, and deploy one of their seismic sensor nodes on the ice there (with support from UIC Science). The IABP maintains a network of Arctic buoys equipped with meteorological and oceanic sensors. Data collected by these buoys are shared with the operational and research communities to support real-time operations (e.g., forecasting sea ice conditions for coastal Alaskans) and climate research. However, these buoys are typically limited in the frequency at which they collect data, so phenomenology on shorter time scales important to climate change may be missed. Moreover, these buoys are difficult and expensive to deploy because they are designed to survive in the harshest environments for years at a time.  The laboratory-developed sensor nodes could offer an inexpensive, easier-to-deploy option for collecting more data over shorter periods of time. In April 2023, Evans placed a sensor node in Utqiaġvik on landfast sea ice, which is stationary ice anchored to the seabed just off the coast. During the sensor node’s week-long deployment, a big piece of drift ice (ice not attached to the seabed or other fixed object) broke off and crashed into the landfast ice. The event was recorded by a radar maintained by the University of Alaska Fairbanks that monitors sea ice movement in near real time to warn of any instability. Though this phenomenology is not exactly the same as that expected for Arctic sea ice, the researchers were encouraged to see seismic activity recorded by their sensor node.In December 2023, Evans and Whelihan headed to New Hampshire, where they conducted echosounder testing in UNH’s engineering test tank and on the Piscataqua River. Together with their UNH partners, they sought to determine whether a low-cost, hobby-grade echosounder could detect the same phenomenology of interest as the high-fidelity UNH echosounder, which would be far too costly to deploy in sensor nodes across the Arctic. In the test tank and on the river, the low-cost echosounder proved capable of detecting masses of water moving in the water column, but with considerably less structural detail than afforded by the higher-cost option. Seeing such dynamics is important to inferring where water comes from and understanding how it affects sea ice breakup — for example, how warm water moving in from the Pacific Ocean is coming into contact with and melting the ice. So, the laboratory researchers and UNH partners have been building a medium-fidelity, medium-cost echosounder.In January 2024, Evans and Whelihan — along with Jehan Diaz, a fellow staff member in their research group — returned to GLRC. With logistical support from their GLRC hosts, they snowmobiled across the ice on Portage Lake, where they practiced several activities to prepare for OIC 2024: augering (drilling) six-inch holes in the ice, albeit in thinner ice than that in the Arctic; placing their long, pipe-like sensor nodes through these holes; operating cold-hardened drones to interact with the nodes; and retrieving the nodes. They also practiced sensor calibration by hitting the ice with an iron bar some distance away from the nodes and correlating this distance with the resulting measured acoustic and seismic intensity.“Our time at GLRC helped us mitigate a lot of risks and prepare to deploy these complex systems in the Arctic,” Whelihan says.Arctic againTo get to OIC, Evans and Whelihan first flew to Prudhoe Bay and reacclimated to the frigid temperatures. They spent the next two days at the Deadhorse Aviation Center hangar inspecting their equipment for transit-induced damage, which included squashed cables and connectors that required rejiggering.“That’s part of the adventure story,” Evans says. “Getting stuff to Prudhoe Bay is not your standard shipping; it’s ice-road trucking.”From there, they boarded a small aircraft to the ice camp.“Even though this trip marked our second time coming here, it was still disorienting,” Evans continues. “You land in the middle of nowhere on a small aircraft after a couple-hour flight. You get out bundled in all of your Arctic gear in this remote, pristine environment.”After unloading and rechecking their equipment for any damage, calibrating their sensors, and attending safety briefings, they were ready to begin their experiments.An icy situationInside the project tent, Evans and Whelihan deployed the UNH-supplied echosounder and a suite of ground-truth sensors on an automated winch to profile water conductivity, temperature, and depth (CTD). Echosounder data needed to be validated with associated CTD data to determine the source of the water in the water column. Ocean properties change as a function of depth, and these changes are important to capture, in part because masses of water coming in from the Atlantic and Pacific oceans arrive at different depths. Though masses of warm water have always existed, climate change–related mechanisms are now bringing them into contact with the ice.  “As ice breaks up, wind can directly interact with the ocean because it’s lacking that barrier of ice cover,” Evans explains. “Kinetic energy from the wind causes mixing in the ocean; all the warm water that used to stay at depth instead gets brought up and interacts with the ice.”They also deployed four of their sensor nodes several miles outside of camp. To access this deployment site, they rode on a sled pulled via a snowmobile driven by Ann Hill, an ASL field party leader trained in Arctic survival and wildlife encounters. The temperature that day was -55 F. At such a dangerously cold temperature, frostnip and frostbite are all too common. To avoid removal of gloves or other protective clothing, the researchers enabled the nodes with WiFi capability (the nodes also have a satellite communications link to transmit low-bandwidth data). Large amounts of data are automatically downloaded over WiFi to an arm-wearable haptic (touch-based) system when a user walks up to a node.“It was so cold that the holes we were drilling in the ice to reach the water column were freezing solid,” Evans explains. “We realized it was going to be quite an ordeal to get our sensor nodes out of the ice.”So, after drilling a big hole in the ice, they deployed only one central node with all the sensor components: a commercial echosounder, an underwater microphone, a seismometer, and a weather station. They deployed the other three nodes, each with a seismometer and weather station, atop the ice.“One of our design considerations was flexibility,” Whelihan says. “Each node can integrate as few or as many sensors as desired.”The small sensor array was only collecting data for about a day when Evans and Whelihan, who were at the time on a helicopter, saw that their initial field site had become completely cut off from camp by a 150-meter-wide ice lead. They quickly returned to camp to load the tools needed to pull the nodes, which were no longer accessible by snowmobile. Two recently arrived staff members from the Ted Stevens Center for Arctic Security Studies offered to help them retrieve their nodes. The helicopter landed on the ice floe near a crack, and the pilot told them they had half an hour to complete their recovery mission. By the time they had retrieved all four sensors, the crack had increased from thumb to fist size.“When we got home, we analyzed the collected sensor data and saw a spike in seismic activity corresponding to what could be the major ice-fracturing event that necessitated our node recovery mission,” Whelihan says.  The researchers also conducted experiments with their Arctic-hardened drones to evaluate their utility for retrieving sensor node data and to develop concepts of operations for future capabilities.“The idea is to have some autonomous vehicle land next to the node, download data, and come back, like a data mule, rather than having to expend energy getting data off the system, say via high-speed satellite communications,” Whelihan says. “We also started testing whether the drone is capable on its own of finding sensors that are constantly moving and getting close enough to them. Even flying in 25-mile-per-hour winds, and at very low temperatures, the drone worked well.”Aside from carrying out their experiments, the researchers had the opportunity to interact with other participants. Their “roommates” were ice scientists from Norway and Finland. They met other ice and water scientists conducting chemistry experiments on the salt content of ice taken from different depths in the ice sheet (when ocean water freezes, salt tends to get pushed out of the ice). One of their collaborators — Nicholas Schmerr, an ice seismologist from the University of Maryland — placed high-quality geophones (for measuring vibrations in the ice) alongside their nodes deployed on the camp field site. They also met with junior enlisted submariners, who temporarily came to camp to open up spots on the submarine for distinguished visitors.“Part of what we’ve been doing over the last three years is building connections within the Arctic community,” Evans says. “Every time I start to get a handle on the phenomenology that exists out here, I learn something new. For example, I didn’t know that sometimes a layer of ice forms a little bit deeper than the primary ice sheet, and you can actually see fish swimming in between the layers.”“One day, we were out with our field party leader, who saw fog while she was looking at the horizon and said the ice was breaking up,” Whelihan adds. “I said, ‘Wait, what?’ As she explained, when an ice lead forms, fog comes out of the ocean. Sure enough, within 30 minutes, we had quarter-mile visibility, whereas beforehand it was unlimited.”Back to solid groundBefore leaving, Whelihan and Evans retrieved and packed up all the remaining sensor nodes, adopting the “leave no trace” philosophy of preserving natural places.“Only a limited number of people get access to this special environment,” Whelihan says. “We hope to grow our footprint at these events in future years, giving opportunities to other laboratory staff members to attend.”In the meantime, they will analyze the collected sensor data and refine their sensor node design. One design consideration is how to replenish the sensors’ battery power. A potential path forward is to leverage the temperature difference between water and air, and harvest energy from the water currents moving under ice floes. Wind energy may provide another viable solution. Solar power would only work for part of the year because the Arctic Circle undergoes periods of complete darkness.The team is also seeking external sponsorship to continue their work engineering sensing systems that advance the scientific community’s understanding of changes to Arctic ice; this work is currently funded through Lincoln Laboratory’s internally administered R&D portfolio on climate change. And, in learning more about this changing environment and its critical importance to strategic interests, they are considering other sensing problems that they could tackle using their Arctic engineering expertise.“The Arctic is becoming a more visible and important region because of how it’s changing,” Evans concludes. “Going forward as a country, we must be able to operate there.” More

    • 113 Shares149 Views

      in Environment

      An AI dataset carves new paths to tornado detection

      by

      The return of spring in the Northern Hemisphere touches off tornado season. A tornado’s twisting funnel of dust and debris seems an unmistakable sight. But that sight can be obscured to radar, the tool of meteorologists. It’s hard to know exactly when a tornado has formed, or even why.

      A new dataset could hold answers. It contains radar returns from thousands of tornadoes that have hit the United States in the past 10 years. Storms that spawned tornadoes are flanked by other severe storms, some with nearly identical conditions, that never did. MIT Lincoln Laboratory researchers who curated the dataset, called TorNet, have now released it open source. They hope to enable breakthroughs in detecting one of nature’s most mysterious and violent phenomena.

      “A lot of progress is driven by easily available, benchmark datasets. We hope TorNet will lay a foundation for machine learning algorithms to both detect and predict tornadoes,” says Mark Veillette, the project’s co-principal investigator with James Kurdzo. Both researchers work in the Air Traffic Control Systems Group. 

      Along with the dataset, the team is releasing models trained on it. The models show promise for machine learning’s ability to spot a twister. Building on this work could open new frontiers for forecasters, helping them provide more accurate warnings that might save lives. 

      Swirling uncertainty

      About 1,200 tornadoes occur in the United States every year, causing millions to billions of dollars in economic damage and claiming 71 lives on average. Last year, one unusually long-lasting tornado killed 17 people and injured at least 165 others along a 59-mile path in Mississippi.  

      Yet tornadoes are notoriously difficult to forecast because scientists don’t have a clear picture of why they form. “We can see two storms that look identical, and one will produce a tornado and one won’t. We don’t fully understand it,” Kurdzo says.

      A tornado’s basic ingredients are thunderstorms with instability caused by rapidly rising warm air and wind shear that causes rotation. Weather radar is the primary tool used to monitor these conditions. But tornadoes lay too low to be detected, even when moderately close to the radar. As the radar beam with a given tilt angle travels further from the antenna, it gets higher above the ground, mostly seeing reflections from rain and hail carried in the “mesocyclone,” the storm’s broad, rotating updraft. A mesocyclone doesn’t always produce a tornado.

      With this limited view, forecasters must decide whether or not to issue a tornado warning. They often err on the side of caution. As a result, the rate of false alarms for tornado warnings is more than 70 percent. “That can lead to boy-who-cried-wolf syndrome,” Kurdzo says.  

      In recent years, researchers have turned to machine learning to better detect and predict tornadoes. However, raw datasets and models have not always been accessible to the broader community, stifling progress. TorNet is filling this gap.

      The dataset contains more than 200,000 radar images, 13,587 of which depict tornadoes. The rest of the images are non-tornadic, taken from storms in one of two categories: randomly selected severe storms or false-alarm storms (those that led a forecaster to issue a warning but that didn’t produce a tornado).

      Each sample of a storm or tornado comprises two sets of six radar images. The two sets correspond to different radar sweep angles. The six images portray different radar data products, such as reflectivity (showing precipitation intensity) or radial velocity (indicating if winds are moving toward or away from the radar).

      A challenge in curating the dataset was first finding tornadoes. Within the corpus of weather radar data, tornadoes are extremely rare events. The team then had to balance those tornado samples with difficult non-tornado samples. If the dataset were too easy, say by comparing tornadoes to snowstorms, an algorithm trained on the data would likely over-classify storms as tornadic.

      “What’s beautiful about a true benchmark dataset is that we’re all working with the same data, with the same level of difficulty, and can compare results,” Veillette says. “It also makes meteorology more accessible to data scientists, and vice versa. It becomes easier for these two parties to work on a common problem.”

      Both researchers represent the progress that can come from cross-collaboration. Veillette is a mathematician and algorithm developer who has long been fascinated by tornadoes. Kurdzo is a meteorologist by training and a signal processing expert. In grad school, he chased tornadoes with custom-built mobile radars, collecting data to analyze in new ways.

      “This dataset also means that a grad student doesn’t have to spend a year or two building a dataset. They can jump right into their research,” Kurdzo says.

      This project was funded by Lincoln Laboratory’s Climate Change Initiative, which aims to leverage the laboratory’s diverse technical strengths to help address climate problems threatening human health and global security.

      Chasing answers with deep learning

      Using the dataset, the researchers developed baseline artificial intelligence (AI) models. They were particularly eager to apply deep learning, a form of machine learning that excels at processing visual data. On its own, deep learning can extract features (key observations that an algorithm uses to make a decision) from images across a dataset. Other machine learning approaches require humans to first manually label features. 

      “We wanted to see if deep learning could rediscover what people normally look for in tornadoes and even identify new things that typically aren’t searched for by forecasters,” Veillette says.

      The results are promising. Their deep learning model performed similar to or better than all tornado-detecting algorithms known in literature. The trained algorithm correctly classified 50 percent of weaker EF-1 tornadoes and over 85 percent of tornadoes rated EF-2 or higher, which make up the most devastating and costly occurrences of these storms.

      They also evaluated two other types of machine-learning models, and one traditional model to compare against. The source code and parameters of all these models are freely available. The models and dataset are also described in a paper submitted to a journal of the American Meteorological Society (AMS). Veillette presented this work at the AMS Annual Meeting in January.

      “The biggest reason for putting our models out there is for the community to improve upon them and do other great things,” Kurdzo says. “The best solution could be a deep learning model, or someone might find that a non-deep learning model is actually better.”

      TorNet could be useful in the weather community for others uses too, such as for conducting large-scale case studies on storms. It could also be augmented with other data sources, like satellite imagery or lightning maps. Fusing multiple types of data could improve the accuracy of machine learning models.

      Taking steps toward operations

      On top of detecting tornadoes, Kurdzo hopes that models might help unravel the science of why they form.

      “As scientists, we see all these precursors to tornadoes — an increase in low-level rotation, a hook echo in reflectivity data, specific differential phase (KDP) foot and differential reflectivity (ZDR) arcs. But how do they all go together? And are there physical manifestations we don’t know about?” he asks.

      Teasing out those answers might be possible with explainable AI. Explainable AI refers to methods that allow a model to provide its reasoning, in a format understandable to humans, of why it came to a certain decision. In this case, these explanations might reveal physical processes that happen before tornadoes. This knowledge could help train forecasters, and models, to recognize the signs sooner. 

      “None of this technology is ever meant to replace a forecaster. But perhaps someday it could guide forecasters’ eyes in complex situations, and give a visual warning to an area predicted to have tornadic activity,” Kurdzo says.

      Such assistance could be especially useful as radar technology improves and future networks potentially grow denser. Data refresh rates in a next-generation radar network are expected to increase from every five minutes to approximately one minute, perhaps faster than forecasters can interpret the new information. Because deep learning can process huge amounts of data quickly, it could be well-suited for monitoring radar returns in real time, alongside humans. Tornadoes can form and disappear in minutes.

      But the path to an operational algorithm is a long road, especially in safety-critical situations, Veillette says. “I think the forecaster community is still, understandably, skeptical of machine learning. One way to establish trust and transparency is to have public benchmark datasets like this one. It’s a first step.”

      The next steps, the team hopes, will be taken by researchers across the world who are inspired by the dataset and energized to build their own algorithms. Those algorithms will in turn go into test beds, where they’ll eventually be shown to forecasters, to start a process of transitioning into operations.

      In the end, the path could circle back to trust.

      “We may never get more than a 10- to 15-minute tornado warning using these tools. But if we could lower the false-alarm rate, we could start to make headway with public perception,” Kurdzo says. “People are going to use those warnings to take the action they need to save their lives.” More

    • 125 Shares119 Views

      in Environment

      Using deep learning to image the Earth’s planetary boundary layer

      by

      Although the troposphere is often thought of as the closest layer of the atmosphere to the Earth’s surface, the planetary boundary layer (PBL) — the lowest layer of the troposphere — is actually the part that most significantly influences weather near the surface. In the 2018 planetary science decadal survey, the PBL was raised as an important scientific issue that has the potential to enhance storm forecasting and improve climate projections.  

      “The PBL is where the surface interacts with the atmosphere, including exchanges of moisture and heat that help lead to severe weather and a changing climate,” says Adam Milstein, a technical staff member in Lincoln Laboratory’s Applied Space Systems Group. “The PBL is also where humans live, and the turbulent movement of aerosols throughout the PBL is important for air quality that influences human health.” 

      Although vital for studying weather and climate, important features of the PBL, such as its height, are difficult to resolve with current technology. In the past four years, Lincoln Laboratory staff have been studying the PBL, focusing on two different tasks: using machine learning to make 3D-scanned profiles of the atmosphere, and resolving the vertical structure of the atmosphere more clearly in order to better predict droughts.  

      This PBL-focused research effort builds on more than a decade of related work on fast, operational neural network algorithms developed by Lincoln Laboratory for NASA missions. These missions include the Time-Resolved Observations of Precipitation structure and storm Intensity with a Constellation of Smallsats (TROPICS) mission as well as Aqua, a satellite that collects data about Earth’s water cycle and observes variables such as ocean temperature, precipitation, and water vapor in the atmosphere. These algorithms retrieve temperature and humidity from the satellite instrument data and have been shown to significantly improve the accuracy and usable global coverage of the observations over previous approaches. For TROPICS, the algorithms help retrieve data that are used to characterize a storm’s rapidly evolving structures in near-real time, and for Aqua, it has helped increase forecasting models, drought monitoring, and fire prediction. 

      These operational algorithms for TROPICS and Aqua are based on classic “shallow” neural networks to maximize speed and simplicity, creating a one-dimensional vertical profile for each spectral measurement collected by the instrument over each location. While this approach has improved observations of the atmosphere down to the surface overall, including the PBL, laboratory staff determined that newer “deep” learning techniques that treat the atmosphere over a region of interest as a three-dimensional image are needed to improve PBL details further.

      “We hypothesized that deep learning and artificial intelligence (AI) techniques could improve on current approaches by incorporating a better statistical representation of 3D temperature and humidity imagery of the atmosphere into the solutions,” Milstein says. “But it took a while to figure out how to create the best dataset — a mix of real and simulated data; we needed to prepare to train these techniques.”

      The team collaborated with Joseph Santanello of the NASA Goddard Space Flight Center and William Blackwell, also of the Applied Space Systems Group, in a recent NASA-funded effort showing that these retrieval algorithms can improve PBL detail, including more accurate determination of the PBL height than the previous state of the art. 

      While improved knowledge of the PBL is broadly useful for increasing understanding of climate and weather, one key application is prediction of droughts. According to a Global Drought Snapshot report released last year, droughts are a pressing planetary issue that the global community needs to address. Lack of humidity near the surface, specifically at the level of the PBL, is the leading indicator of drought. While previous studies using remote-sensing techniques have examined the humidity of soil to determine drought risk, studying the atmosphere can help predict when droughts will happen.  

      In an effort funded by Lincoln Laboratory’s Climate Change Initiative, Milstein, along with laboratory staff member Michael Pieper, are working with scientists at NASA’s Jet Propulsion Laboratory (JPL) to use neural network techniques to improve drought prediction over the continental United States. While the work builds off of existing operational work JPL has done incorporating (in part) the laboratory’s operational “shallow” neural network approach for Aqua, the team believes that this work and the PBL-focused deep learning research work can be combined to further improve the accuracy of drought prediction. 

      “Lincoln Laboratory has been working with NASA for more than a decade on neural network algorithms for estimating temperature and humidity in the atmosphere from space-borne infrared and microwave instruments, including those on the Aqua spacecraft,” Milstein says. “Over that time, we have learned a lot about this problem by working with the science community, including learning about what scientific challenges remain. Our long experience working on this type of remote sensing with NASA scientists, as well as our experience with using neural network techniques, gave us a unique perspective.”

      According to Milstein, the next step for this project is to compare the deep learning results to datasets from the National Oceanic and Atmospheric Administration, NASA, and the Department of Energy collected directly in the PBL using radiosondes, a type of instrument flown on a weather balloon. “These direct measurements can be considered a kind of ‘ground truth’ to quantify the accuracy of the techniques we have developed,” Milstein says.

      This improved neural network approach holds promise to demonstrate drought prediction that can exceed the capabilities of existing indicators, Milstein says, and to be a tool that scientists can rely on for decades to come. More

    • 75 Shares169 Views

      in Energy

      Generative AI for smart grid modeling

      by

      MIT’s Laboratory for Information and Decision Systems (LIDS) has been awarded $1,365,000 in funding from the Appalachian Regional Commission (ARC) to support its involvement with an innovative project, “Forming the Smart Grid Deployment Consortium (SGDC) and Expanding the HILLTOP+ Platform.”

      The grant was made available through ARC’s Appalachian Regional Initiative for Stronger Economies, which fosters regional economic transformation through multi-state collaboration.

      Led by Kalyan Veeramachaneni, research scientist and principal investigator at LIDS’ Data to AI Group, the project will focus on creating AI-driven generative models for customer load data. Veeramachaneni and colleagues will work alongside a team of universities and organizations led by Tennessee Tech University, including collaborators across Ohio, Pennsylvania, West Virginia, and Tennessee, to develop and deploy smart grid modeling services through the SGDC project.

      These generative models have far-reaching applications, including grid modeling and training algorithms for energy tech startups. When the models are trained on existing data, they create additional, realistic data that can augment limited datasets or stand in for sensitive ones. Stakeholders can then use these models to understand and plan for specific what-if scenarios far beyond what could be achieved with existing data alone. For example, generated data can predict the potential load on the grid if an additional 1,000 households were to adopt solar technologies, how that load might change throughout the day, and similar contingencies vital to future planning.

      The generative AI models developed by Veeramachaneni and his team will provide inputs to modeling services based on the HILLTOP+ microgrid simulation platform, originally prototyped by MIT Lincoln Laboratory. HILLTOP+ will be used to model and test new smart grid technologies in a virtual “safe space,” providing rural electric utilities with increased confidence in deploying smart grid technologies, including utility-scale battery storage. Energy tech startups will also benefit from HILLTOP+ grid modeling services, enabling them to develop and virtually test their smart grid hardware and software products for scalability and interoperability.

      The project aims to assist rural electric utilities and energy tech startups in mitigating the risks associated with deploying these new technologies. “This project is a powerful example of how generative AI can transform a sector — in this case, the energy sector,” says Veeramachaneni. “In order to be useful, generative AI technologies and their development have to be closely integrated with domain expertise. I am thrilled to be collaborating with experts in grid modeling, and working alongside them to integrate the latest and greatest from my research group and push the boundaries of these technologies.”

      “This project is testament to the power of collaboration and innovation, and we look forward to working with our collaborators to drive positive change in the energy sector,” says Satish Mahajan, principal investigator for the project at Tennessee Tech and a professor of electrical and computer engineering. Tennessee Tech’s Center for Rural Innovation director, Michael Aikens, adds, “Together, we are taking significant steps towards a more sustainable and resilient future for the Appalachian region.” More

    • 163 Shares169 Views

      in Energy

      New tools are available to help reduce the energy that AI models devour

      by

      When searching for flights on Google, you may have noticed that each flight’s carbon-emission estimate is now presented next to its cost. It’s a way to inform customers about their environmental impact, and to let them factor this information into their decision-making.

      A similar kind of transparency doesn’t yet exist for the computing industry, despite its carbon emissions exceeding those of the entire airline industry. Escalating this energy demand are artificial intelligence models. Huge, popular models like ChatGPT signal a trend of large-scale artificial intelligence, boosting forecasts that predict data centers will draw up to 21 percent of the world’s electricity supply by 2030.

      The MIT Lincoln Laboratory Supercomputing Center (LLSC) is developing techniques to help data centers reel in energy use. Their techniques range from simple but effective changes, like power-capping hardware, to adopting novel tools that can stop AI training early on. Crucially, they have found that these techniques have a minimal impact on model performance.

      In the wider picture, their work is mobilizing green-computing research and promoting a culture of transparency. “Energy-aware computing is not really a research area, because everyone’s been holding on to their data,” says Vijay Gadepally, senior staff in the LLSC who leads energy-aware research efforts. “Somebody has to start, and we’re hoping others will follow.”

      Curbing power and cooling down

      Like many data centers, the LLSC has seen a significant uptick in the number of AI jobs running on its hardware. Noticing an increase in energy usage, computer scientists at the LLSC were curious about ways to run jobs more efficiently. Green computing is a principle of the center, which is powered entirely by carbon-free energy.

      Training an AI model — the process by which it learns patterns from huge datasets — requires using graphics processing units (GPUs), which are power-hungry hardware. As one example, the GPUs that trained GPT-3 (the precursor to ChatGPT) are estimated to have consumed 1,300 megawatt-hours of electricity, roughly equal to that used by 1,450 average U.S. households per month.

      While most people seek out GPUs because of their computational power, manufacturers offer ways to limit the amount of power a GPU is allowed to draw. “We studied the effects of capping power and found that we could reduce energy consumption by about 12 percent to 15 percent, depending on the model,” Siddharth Samsi, a researcher within the LLSC, says.

      The trade-off for capping power is increasing task time — GPUs will take about 3 percent longer to complete a task, an increase Gadepally says is “barely noticeable” considering that models are often trained over days or even months. In one of their experiments in which they trained the popular BERT language model, limiting GPU power to 150 watts saw a two-hour increase in training time (from 80 to 82 hours) but saved the equivalent of a U.S. household’s week of energy.

      The team then built software that plugs this power-capping capability into the widely used scheduler system, Slurm. The software lets data center owners set limits across their system or on a job-by-job basis.

      “We can deploy this intervention today, and we’ve done so across all our systems,” Gadepally says.

      Side benefits have arisen, too. Since putting power constraints in place, the GPUs on LLSC supercomputers have been running about 30 degrees Fahrenheit cooler and at a more consistent temperature, reducing stress on the cooling system. Running the hardware cooler can potentially also increase reliability and service lifetime. They can now consider delaying the purchase of new hardware — reducing the center’s “embodied carbon,” or the emissions created through the manufacturing of equipment — until the efficiencies gained by using new hardware offset this aspect of the carbon footprint. They’re also finding ways to cut down on cooling needs by strategically scheduling jobs to run at night and during the winter months.

      “Data centers can use these easy-to-implement approaches today to increase efficiencies, without requiring modifications to code or infrastructure,” Gadepally says.

      Taking this holistic look at a data center’s operations to find opportunities to cut down can be time-intensive. To make this process easier for others, the team — in collaboration with Professor Devesh Tiwari and Baolin Li at Northeastern University — recently developed and published a comprehensive framework for analyzing the carbon footprint of high-performance computing systems. System practitioners can use this analysis framework to gain a better understanding of how sustainable their current system is and consider changes for next-generation systems.  

      Adjusting how models are trained and used

      On top of making adjustments to data center operations, the team is devising ways to make AI-model development more efficient.

      When training models, AI developers often focus on improving accuracy, and they build upon previous models as a starting point. To achieve the desired output, they have to figure out what parameters to use, and getting it right can take testing thousands of configurations. This process, called hyperparameter optimization, is one area LLSC researchers have found ripe for cutting down energy waste. 

      “We’ve developed a model that basically looks at the rate at which a given configuration is learning,” Gadepally says. Given that rate, their model predicts the likely performance. Underperforming models are stopped early. “We can give you a very accurate estimate early on that the best model will be in this top 10 of 100 models running,” he says.

      In their studies, this early stopping led to dramatic savings: an 80 percent reduction in the energy used for model training. They’ve applied this technique to models developed for computer vision, natural language processing, and material design applications.

      “In my opinion, this technique has the biggest potential for advancing the way AI models are trained,” Gadepally says.

      Training is just one part of an AI model’s emissions. The largest contributor to emissions over time is model inference, or the process of running the model live, like when a user chats with ChatGPT. To respond quickly, these models use redundant hardware, running all the time, waiting for a user to ask a question.

      One way to improve inference efficiency is to use the most appropriate hardware. Also with Northeastern University, the team created an optimizer that matches a model with the most carbon-efficient mix of hardware, such as high-power GPUs for the computationally intense parts of inference and low-power central processing units (CPUs) for the less-demanding aspects. This work recently won the best paper award at the International ACM Symposium on High-Performance Parallel and Distributed Computing.

      Using this optimizer can decrease energy use by 10-20 percent while still meeting the same “quality-of-service target” (how quickly the model can respond).

      This tool is especially helpful for cloud customers, who lease systems from data centers and must select hardware from among thousands of options. “Most customers overestimate what they need; they choose over-capable hardware just because they don’t know any better,” Gadepally says.

      Growing green-computing awareness

      The energy saved by implementing these interventions also reduces the associated costs of developing AI, often by a one-to-one ratio. In fact, cost is usually used as a proxy for energy consumption. Given these savings, why aren’t more data centers investing in green techniques?

      “I think it’s a bit of an incentive-misalignment problem,” Samsi says. “There’s been such a race to build bigger and better models that almost every secondary consideration has been put aside.”

      They point out that while some data centers buy renewable-energy credits, these renewables aren’t enough to cover the growing energy demands. The majority of electricity powering data centers comes from fossil fuels, and water used for cooling is contributing to stressed watersheds. 

      Hesitancy may also exist because systematic studies on energy-saving techniques haven’t been conducted. That’s why the team has been pushing their research in peer-reviewed venues in addition to open-source repositories. Some big industry players, like Google DeepMind, have applied machine learning to increase data center efficiency but have not made their work available for others to deploy or replicate. 

      Top AI conferences are now pushing for ethics statements that consider how AI could be misused. The team sees the climate aspect as an AI ethics topic that has not yet been given much attention, but this also appears to be slowly changing. Some researchers are now disclosing the carbon footprint of training the latest models, and industry is showing a shift in energy transparency too, as in this recent report from Meta AI.

      They also acknowledge that transparency is difficult without tools that can show AI developers their consumption. Reporting is on the LLSC roadmap for this year. They want to be able to show every LLSC user, for every job, how much energy they consume and how this amount compares to others, similar to home energy reports.

      Part of this effort requires working more closely with hardware manufacturers to make getting these data off hardware easier and more accurate. If manufacturers can standardize the way the data are read out, then energy-saving and reporting tools can be applied across different hardware platforms. A collaboration is underway between the LLSC researchers and Intel to work on this very problem.

      Even for AI developers who are aware of the intense energy needs of AI, they can’t do much on their own to curb this energy use. The LLSC team wants to help other data centers apply these interventions and provide users with energy-aware options. Their first partnership is with the U.S. Air Force, a sponsor of this research, which operates thousands of data centers. Applying these techniques can make a significant dent in their energy consumption and cost.

      “We’re putting control into the hands of AI developers who want to lessen their footprint,” Gadepally says. “Do I really need to gratuitously train unpromising models? Am I willing to run my GPUs slower to save energy? To our knowledge, no other supercomputing center is letting you consider these options. Using our tools, today, you get to decide.”

      Visit this webpage to see the group’s publications related to energy-aware computing and findings described in this article. More