Can AI Make a Better Fusion Reactor?

Nuclear physics may be one of machine learning's newest frontiers

3 min read
Spherical ball of plasma (pink) in a red colored chamber
EFDA-JET/Science Source

Since the 1940s, physicists have tried, but no one has yet created an efficient nuclear fusion reaction. Meanwhile, AI and machine learning (ML) have, across many industries and applications, proved themselves quite capable at detecting subtle patterns in data that humans can't recognize. So could neural nets and the GPUs that power them help in nuclear fusion? The challenge, and it's a big one, would be to accelerate the worldwide quest to tame instabilities in hot plasmas and ultimately provide a source of sustainable, and carbon-free power.

"Physicists, they develop theoretical models, they write equations, they manipulate things mathematically," said Diogo Ferreira, a professor of information systems at the University of Lisbon's Instituto Superior Técnico in Portugal. "But there is a limit to that." AI, he says, can help.

Ferreira recently collaborated with colleagues working on the Joint European Torus (JET) in the UK in a study that detailed three different uses for AI, machine learning, and deep learning models for fusion research. Ferreira trained his models using diagnostic data from 48 sensors connected to the JET reactor, called bolometers, which collect power and radiation data.

One of Ferreira's models predicts disruptions in a super-hot plasma. In the study, he explains that depending on how it is trained, the model can either predict the likelihood of disruption—which can result in a plasma escaping confinement, jolting equipment, drastically reducing the plasma's temperature, and ending the reaction— or estimate the time at which that disruption will occur.

A second model detects anomalies in the plasma. Trained only on reactions where disruptions did not occur, the model can reproduce these "good" experiments. If the data originates in an experiment that ended in a disruption, the model can identify when and how the data diverges from that of a successful reaction. Scientists could use this process to better understand what ultimately leads to disruptions and eventually to run reactions in which disruptions are less likely.

Another application concerns visual representations of plasma radiation patterns. Performing brute-force, direct calculations, Ferriera says, can take 20 minutes for each reaction. By contrast, another model from Ferreira's research group can produce similar images in seconds or even less. It's so fast, Ferreira says it could one day be done during an experiment in real-time.

Researchers at the University of Washington, including Kyle Morgan and Chris Hansen, recently published a study detailing a method that uses machine learning to predict the behavior of a plasma. Their model, which uses a statistical technique called regression, essentially throws out scenarios that lead to nonsensical results, enabling it to use less data, less computational power, and less time. Hansen says that although the model in the study doesn't work quickly enough to use during an experiment, he thinks that it eventually could. The researchers published another recent study that used a single GPU to control a fusion experiment that had previously required several computers. This kind of powerful system, Hansen says, could eventually be used to run the model quickly enough that it would be useful during an experiment.

Other methods can be used before or after an experiment. In a recent study, Stefano Markidis, an associate professor of computer science at KTH Royal Institute of Technology in Stockholm, Sweden, along with his colleague Xavier Aguilar, created a deep learning model that solves one of the more computationally intensive steps of determining information on a plasma–calculating its electric field. The method was faster, and in some cases, more accurate than the traditional method involving complex mathematical equations.

AI and machine learning are not without their disadvantages in nuclear fusion systems. Machine learning algorithms, especially deep learning models, are "black boxes"–it isn't always possible to know how a model gets its results. But by working with these algorithms, scientists can glean bits of what these models see, and learn more about the physics of plasma and fusion.

"At the end of the day, it's going to be our minds that [will] solve the fusion problem," he says. "It's just a matter of what tools we use, and AI and machine learning will be a key tool."

The Conversation (1)
Frederic Andre 26 Aug, 2021
M

There are a lot of claims in this article like "Ferreira's models predicts disruptions in a super-hot plasma". Are they really sustained scientifically ?

Smokey the AI

Smart image analysis algorithms, fed by cameras carried by drones and ground vehicles, can help power companies prevent forest fires

7 min read
Smokey the AI

The 2021 Dixie Fire in northern California is suspected of being caused by Pacific Gas & Electric's equipment. The fire is the second-largest in California history.

Robyn Beck/AFP/Getty Images

The 2020 fire season in the United States was the worst in at least 70 years, with some 4 million hectares burned on the west coast alone. These West Coast fires killed at least 37 people, destroyed hundreds of structures, caused nearly US $20 billion in damage, and filled the air with smoke that threatened the health of millions of people. And this was on top of a 2018 fire season that burned more than 700,000 hectares of land in California, and a 2019-to-2020 wildfire season in Australia that torched nearly 18 million hectares.

While some of these fires started from human carelessness—or arson—far too many were sparked and spread by the electrical power infrastructure and power lines. The California Department of Forestry and Fire Protection (Cal Fire) calculates that nearly 100,000 burned hectares of those 2018 California fires were the fault of the electric power infrastructure, including the devastating Camp Fire, which wiped out most of the town of Paradise. And in July of this year, Pacific Gas & Electric indicated that blown fuses on one of its utility poles may have sparked the Dixie Fire, which burned nearly 400,000 hectares.

Until these recent disasters, most people, even those living in vulnerable areas, didn't give much thought to the fire risk from the electrical infrastructure. Power companies trim trees and inspect lines on a regular—if not particularly frequent—basis.

However, the frequency of these inspections has changed little over the years, even though climate change is causing drier and hotter weather conditions that lead up to more intense wildfires. In addition, many key electrical components are beyond their shelf lives, including insulators, transformers, arrestors, and splices that are more than 40 years old. Many transmission towers, most built for a 40-year lifespan, are entering their final decade.

Keep Reading ↓ Show less