A groundbreaking study from Texas A&M University has led to the creation of a Power System Vulnerability Index using AI. This tool highlights U.S. regions most susceptible to power outages, aiming to help policymakers bolster infrastructure.
Amid the increasing threat of severe weather events like Hurricane Beryl and Winter Storm Uri, long-term power outages have become a significant concern. Texas residents, in particular, have faced frequent disruptions, but a newly developed tool from Texas A&M University aims to address this issue on a national scale.
Researchers at the Urban Resilience AI Lab at Texas A&M have utilized machine learning to create a Power System Vulnerability Index (PSVI), a nationwide tool identifying regions at an elevated risk of power outages.
“Using data from Oak Ridge National Laboratory, we were able to study the effect of weather events on the frequency and duration of nationwide power outages over the past 10 years,” Junwei Ma, a postdoctoral researcher in the Zachry Department of Civil and Environmental Engineering, said in a news release. “The dataset included over 179 million data points sorted by time and location, allowing us to create the PSVI.”
The study, published in the journal Applied Energy, reveals a troubling trend: a noticeable increase in both the frequency and duration of weather-induced power outages across the United States.
Ma, along with his fellow postdoctoral researcher Bo Li, Olufemi A. Omitaomu from Oak Ridge National Laboratory, and Ali Mostafavi, a professor in the Zachry Department, pinpointed several high-risk areas.
The East and West Coasts, as well as the Great Lakes and Gulf regions, emerged as hot spots with significant power system vulnerabilities.
“This is an interactive tool that can showcase the overall PSVI ratings and scores of individual U.S. counties over the past decade, and how vulnerability shifts year by year,” added Ma.
The publicly available PSVI map provides a clear visual representation of areas at risk, thus equipping policymakers with critical information for infrastructure planning and investment.
Investigating further, the research team noticed many AI data centers, crucial for storing massive amounts of data, are located in these identified hot spots.
This finding underscores the necessity for reinforced infrastructure to protect vital digital resources.
Mostafavi, who also serves as the director of the Urban Resilience AI Lab, elaborated on the broader implications, adding: “We knew that the state of power system vulnerability nationwide is exacerbating. But the magnitude of that was shocking, and greater than we hypothesized. After 2019, we see a 20% annual increase in outage duration, frequency and magnitude.”
This increase in power system vulnerability is more than just an inconvenience; it has significant socioeconomic impacts. Long and frequent power outages can limit access to food, hinder transportation and affect daily life.
Knowing which areas are most vulnerable allows decision makers to prioritize resource allocation and infrastructure improvements, thereby enhancing community resilience.
Source: Texas A&M University