Transformer shortages loom for strained US power grid
As data centers and electric vehicles increase power demand, the country could run out of transformers and other equipment as early as 2027, new study finds
image:
If high power demand continues as expected, over the next four years the researchers project the U.S. will be almost 15% short on transformers and 30% short on other critical equipment.
view moreCredit: Boyu Yao,/Johns Hopkins University
As growth in artificial intelligence forces the United States to expand its power supply, a new study finds the country will soon hit a surprising roadblock: shortages of key materials needed to connect new data centers and power plants to the grid.
In a first-of-its-kind study, Johns Hopkins University researchers found the country could run short of critical grid supporting equipment, such as transformers and converters, as early as next year, with acute shortages hitting by 2030.
“This type of equipment was taken for granted but without it, you can build a data center or a power plant but it won’t be powered because you cannot connect it to the grid,” said author Yury Dvorkin, a researcher with expertise in modeling power and infrastructure demand. “When the demand for electricity started to grow faster than expected, it started to amplify all the cracks in the industry.”
The work published today in Cell Reports Sustainability.
Historically the country’s power demand has grown incrementally, at a percent or two a year. But thanks to power-hungry data centers, growth in electric vehicles, and other needs, power use is projected to grow 21% by 2030.
The ability to grow the U.S. power grid and to refurbish aging infrastructure depends on the availability of key equipment, including, most importantly, power transformers and transformers for data centers. There are also things like charging equipment for electric vehicles, and power electronics converters for wind power, batteries and solar panels.
The materials needed to build transformers and other grid-supporting equipment include copper, steel, nickel, aluminum and manganese. The U.S. is in danger of running out of all of them over the next four years, the Johns Hopkins study found.
Dvorkin and Boyu Yao, a PhD candidate in Civil and Systems Engineering who studies the challenges for evolving power grids, modeled power demand projections alongside global supply chain data to determine whether the country would have the materials needed to support data center growth in coming years. The study is the first to precisely quantify expected shortages by equipment type and material.
The team projects the U.S. will run short of copper next year, effecting not only equipment for data centers, but also converters for wind power generation and electric vehicle charging infrastructure.
Steel and nickel will be nearly impossible to source by 2030, the team found.
Under baseline scenarios, or modest growth in power demand, by 2030 the nation would likely have enough transformers, but fall about 16% short of other grid-supporting equipment.
But if high power demand continues as expected, over the next four years the researchers project the U.S. will be almost 15% short on transformers and 30% short on other critical equipment.
As much as two-thirds of all the expected equipment shortages will be fueled by data centers, according to the researchers.
If international trade disruptions occur, on top of the increases in demand, the result could be as much as a 70% drop in materials coming in from unstable trade partners, leading to a transformer shortage of about 25%, and a 36% shortage for other equipment, the team found.
Already lead times for transformers, which are typically custom-built rather than mass-produced, have quadrupled over the last four years, from six months to more than two years, Dvorkin said. Compounding the issue, existing transformers that are aging out must be refurbished or replaced.
“By 2030 we’re likely to have transformer shortages even without the most aggressive power demand growth,” Dvorkin said.
Dvorkin said individuals are likely to feel the effects of the infrastructure bottlenecks through higher electric bills and more frequent power outages as power companies compete with other corporations for the scarce transformers.
“Utilities will compete for transformers with the data center industry,” Dvorkin said. “Transformers will first be delivered to whoever can pay the highest premium. Will it be a mom-and-pop utility or a big data center corporation?”
Using low-cost sensors to monitor and adjust transformer load in real time could slightly ease the shortages by tracking actual temperature and load, rather than relying on conservative fixed ratings, so utilities could get more out of the transformers they already have and replace them less often while reducing overall demand, the team said. However, that technology is currently not widely used. Another opportunity to alleviate the peak demand is to develop methods to recycle existing transformer materials at the end of their lifetime, along with standardization efforts that will help improve transformer substitutability, the team said.
Journal
Cell Reports Sustainability
The team projects the U.S. will run short of copper next year, effecting not only equipment for data centers, but also converters for wind power generation and electric vehicle charging infrastructure. Steel and nickel will be nearly impossible to source by 2030, the team found.
Credit
Boyu Yao/Johns Hopkins University
How insider threats reshape cybersecurity risks in power grids
KeAi Communications Co., Ltd.
Cybersecurity threats to power grids do not always come from outside. People with legitimate access to power systems can also create pathways for attacks, whether through malicious actions or by assisting external attackers. As information technology (IT) and operational technology (OT) become more closely connected, these trusted access pathways can expose critical operations to risks that conventional perimeter defences may not adequately address.
A study published in Cyber-Physical Energy Systems examines how insider and external threats can affect power transmission systems differently. The study’s researchers from Michigan Technological University and the University of Michigan-Dearborn found that the key distinction is not simply how an attack begins, but how easily it can progress through the systems and organisations that support grid operations. The study’s framework offers a way to assess these interacting risks by estimating the long-run probabilities of different system states and considering their potential consequences. It could complement established requirements, including the North American Electric Reliability Corporation Critical Infrastructure Protection standards, by providing a probabilistic perspective on risk alongside compliance-based evaluation.
“External attackers may need to overcome multiple barriers to reach sensitive operational systems,” explains Roman Sidortsov. “Insiders, or attackers who gain assistance from people with legitimate access, may be able to exploit existing privileges and knowledge of system operations to bypass some of these barriers.”
In the researchers' model, such attacks can reach high-impact system states through fewer, more probable transitions. This means that fewer observable attack steps do not necessarily indicate lower risk.
The team's findings highlighted a limitation of relying on perimeter protection and procedural compliance alone. “Cybersecurity risks also depend on how access is granted and used, how organisations manage trusted relationships, and how technical systems are connected,” says Sidortsov. “An attack that begins with a compromised account or trusted access pathway may have consequences beyond the initially affected system as it moves across interconnected IT and OT environments.”
To examine how different safeguards affect these risks, the researchers simulated several cybersecurity measures, including network segmentation, privileged access workstations, intrusion detection and security information and event management, and secure firmware validation. “Under balanced risk weighting, zero-trust segmentation achieved the lowest normalised defender objective among the evaluated policies, improving on the baseline by 7.6%,” shares Lawrence Dilworth. “It also reduced modelled cyber-compromise exposure by 68.55%. Privileged access workstations had a different benefit, reducing modelled cyber-physical coupling exposure by 45.58%.”
However, no single measure substantially reduced all the major risk dimensions considered in the simulations. Some measures lowered the likelihood of cyber compromise while leaving greater residual risk associated with the propagation of attacks into more consequential states. The preferred measure also varied with the assumed mix of external, insider and supply-chain threats.
“The results suggest that protecting power grids requires more than selecting one leading security control: safeguards need to address different stages of an attack, from restricting access and lateral movement to detecting intrusions, containing their spread and supporting recovery,” says Dilworth.
“The simulations illustrate how the effectiveness of cybersecurity measures can depend on who is attacking and how an attack may progress,” adds Sidortsov. “For power grid operators, the central implication is that trusted access deserves explicit attention in security planning. A layered approach that combines access controls, network segmentation, monitoring and recovery measures is better suited to addressing varied threat pathways than reliance on perimeter defences or procedural compliance alone.”
Contact the Authors:
Chee-Wooi Ten*, Professor, Department of Electrical and Computer Engineering, Michigan Technological University, Houghton, MI 49931, USA, ten@mtu.edu
Lawrence Dilworth, Department of Electrical and Computer Engineering, Michigan Technological University, Houghton, MI 49931, USA
Junho Hong, Associate Professor, Department of Electrical and Computer Engineering, University of Michigan-Dearborn, Dearborn, MI 48128, USA
Swetha Rani Kasimalla, Department of Electrical and Computer Engineering, University of Michigan-Dearborn, Dearborn, MI 48128, USA
Roman Sidortsov, Associate Professor, Department of Social Sciences, Michigan Technological University, Houghton, MI 49931, USA
The publisher KeAi was established by Elsevier and China Science Publishing & Media Ltd to unfold quality research globally. In 2013, our focus shifted to open access publishing. We now proudly publish more than 300 world-class, open access, English language journals, spanning all scientific disciplines. Many of these are titles we publish in partnership with prestigious societies and academic institutions, such as the National Natural Science Foundation of China (NSFC).
Journal
Cyber-Physical Energy Systems
Article Title
The emerging landscape of cyberthreats in SCADA deployment
No comments:
Post a Comment