At Indiana’s only nuclear reactor, engineers conduct the first experiments of their kind in the U.S.

Underground on Purdue University’s campus is the only nuclear reactor in Indiana.

Although used just for research purposes — the total energy the reactor generates powers about the equivalent of 10 microwaves — Purdue University Reactor Number One has specific features that no other reactor in the U.S. has. But those features are coming in the next generation of reactors.

With these features, future reactors could cost less to operate and maintain, be safer and last longer — removing barriers in building additional reactors to increase generation of carbon-free electricity. Research that Purdue engineers are conducting with these features is helping develop new techniques that reactors could use to achieve these goals, which align with the university’s efforts to investigate nuclear energy.

These features could be summed up with one word: “digital.” Purdue’s reactor, called “PUR-1” for short, is the first in the nation to be controlled and operated digitally — think computer screens, keyboards and ethernet cables — rather than with dials, knobs and other analog technology that U.S. reactors have been using since the 1960s.

Although some countries already have reactors with digital controls, PUR-1 is the only all-digital reactor that has been licensed by the U.S. Nuclear Regulatory Commission. All-digital means that the “nervous system” of the reactor, its instrumentation and control system, entirely uses digital technology. The digital capabilities of other reactors in the U.S. are mostly limited to sensors and have not been applied to controls.

“Our switch to digital instrumentation and control signaled to the nuclear industry that this is possible in the U.S.,” said Seungjin Kim, the Capt. James F. McCarthy, Jr. and Cheryl E. McCarthy Head of Purdue’s School of Nuclear Engineering and facility director of PUR-1.

PUR-1 was built in 1962 and converted from analog to digital in 2019 with support from the Department of Energy’s Office of Nuclear Energy (grant DE-NE000498). Since this digital upgrade, Purdue engineering faculty and students have been performing first-of-a-kind experiments that are unique to the nuclear sector.

Their findings are helping inform the development of advanced reactors such as small modular reactors and microreactors, which would be significantly smaller and easier to construct than existing reactors so that they can power more communities, even in rural or remote areas. For efficiency, many of these reactors will be operated from a distance by the same control center, which means they will need to communicate digitally.

Going digital would also allow operators to take measurements from a reactor in real time and use artificial intelligence tools to monitor the reactor’s performance. They could better predict and detect problems in between regularly scheduled maintenance, which would improve a reactor’s safety and lifespan.

Revealing the potential of AI for nuclear reactors

PUR-1 has started serving as the nation’s first reactor test bed to help the industry figure out how digital communication, AI tools and cybersecurity methods could work at a larger scale for advanced reactors.

The lab of Purdue nuclear engineering assistant professor and associate PUR-1 director Stylianos Chatzidakis completed building a “digital twin” of PUR-1 in 2023 that has allowed his research group and collaborators to do experiments on a digital copy of the reactor without affecting its operation. Funding from the Department of Energy’s Office of Nuclear Energy supported the development of the digital twin (grant DE-NE0009174 and DE-NE0009268).

The digital twin is a fully integrated physics and data-driven simulation that receives measurements in real time from PUR-1’s sensors, makes predictions using AI-driven algorithms and provides insights that can inform reactor operations. Chatzidakis and his students access the twin on computers in a lab adjacent to the reactor facility.

“We are the only university that has a digital twin of a true nuclear reactor that can utilize reactor-generated signals for research. That makes us unique,” Kim said.

In a study published in Nature’s Scientific Reports, Chatzidakis and other collaborators from Purdue and Argonne National Laboratory showed how PUR-1’s digital twin could test a machine learning algorithm they developed for improving the performance of small modular reactors. They found that the algorithm could rapidly learn about the physics behind a measurement of how steadily the reactor is producing power and predict changes in this indicator over time with 99% accuracy.

Being able to access PUR-1’s measurements from a different building has made it possible for Chatzidakis’ lab to explore how a similar framework might work in the future to monitor and operate advanced reactors from remote locations.

“Let’s say that you have a fleet of small modular reactors or microreactors operating in a remote location. If staff could be in a control room hundreds or thousands of miles away and monitor multiple reactors at once, we could minimize the operation and maintenance costs. Using PUR-1, we could quantify the potential reduction in costs,” Chatzidakis said.

Thinking ahead on cybersecurity needs as reactors advance

But to remotely operate reactors, communications would need to be secure from potential cyberattacks. In a technical letter report published by the U.S. Nuclear Regulatory Commission, Chatzidakis and other Purdue researchers conducted a project using real-time reactor data to evaluate how various AI and machine learning models could distinguish abnormal from normal cybersecurity states within nuclear systems.

The real-time data, available through PUR-1, helped train and test models in one of the cybersecurity use cases demonstrated in this project. Results showed that artificial intelligence and machine learning models could successfully detect abnormal cybersecurity events. “The idea is that the nuclear industry could refer to this report as they develop machine learning for cybersecurity,” Chatzidakis said.

Chatzidakis’ lab also has been using PUR-1 to study the feasibility of using quantum encryption to protect communications coming in or out of a reactor.

“Encryption based on quantum principles cannot be broken with any computer. It doesn’t matter if you have a supercomputer or a quantum computer — it’s unbreakable,” he said.

With data from PUR-1, Chatzidakis and his students have simulated how quantum encryption might work to remotely monitor and operate advanced reactors. Next, they plan to do experiments and gather real-world data to test if quantum equipment can encrypt signals from PUR-1, which they will access via the digital twin.