In a groundbreaking development, researchers at Brigham Young University (BYU) have made significant strides in nuclear power design by leveraging artificial intelligence. Led by Professor Matthew Memmott, a team from BYU's Department of Chemical Engineering has developed an AI algorithm that promises to make the nuclear reactor design process faster, more cost-effective, and safer—a true game-changer in the energy sector.
Designing a nuclear reactor is an extraordinarily complex and expensive endeavor. Before construction can even begin, the proposed design must receive approval from the U.S. Nuclear Regulatory Commission (NRC). This approval process can take decades and cost upwards of a billion dollars, as engineers spend years meticulously calculating every aspect of the reactor’s design to ensure it meets rigorous safety standards.
Professor Memmott highlights the enormity of this challenge: "A nuclear reactor cannot be built until it receives approval, so a team of engineers must work on calculations for around 20 years and spend about a billion dollars just to get that license to build it." The need for innovation in this space is clear, and that’s where Memmott and his team have turned to machine learning for a solution.
Memmott and his team have developed a machine learning algorithm that dramatically accelerates the design process. “The idea is to shorten it, make it safer, cheaper, and faster to get the nuclear power, rather than take 20 years to get the license,” Memmott explains. By utilizing AI to process vast amounts of data and perform complex calculations in a fraction of the time it would take a human team, the researchers are transforming the way nuclear reactors are designed.
To test their approach, the BYU team recreated a local nuclear company’s shield design using their AI algorithm. Remarkably, the AI-produced design was nearly identical to the original, and it achieved this in just two days—work that previously took a team of engineers six months. Although the AI-generated shield was slightly less effective, it was also cheaper and lighter, demonstrating the potential for AI to optimize nuclear designs without compromising on quality.
Memmott recounts the company’s reaction: “They were like, this is amazing because within a couple of days we were able to do the same work that it took a team of engineers six months to do.”
The implications of this research are profound. By drastically reducing the time and cost required to design and license nuclear reactors, AI could help usher in a new era of nuclear power—one that is more accessible, more affordable, and more aligned with the global push for clean, sustainable energy.
Graduate student Edward Mercado, who assisted in running simulation models for the project, is optimistic about the future. “Machine learning has proved to be a very valuable tool for design and optimization,” Mercado says. “My hope is that a similar machine learning framework can be used to optimize nuclear fuel consumption so that less waste is generated.”
Indeed, the team’s research, recently published in a paper on using AI to optimize nuclear shielding, shows that AI can significantly enhance the efficiency and cost-effectiveness of nuclear power design. But Memmott believes the potential applications of their AI algorithm extend beyond nuclear reactors.
### Expanding the Scope of AI in Energy
The research team is already exploring other areas where their AI algorithm could be applied. One exciting prospect is optimizing renewable energy production within electricity grids. “Applying AI to other areas is something we’re going to expand to see if we can figure out what else can be improved,” Memmott says.
Additionally, the team is collaborating with the Pacific Northwest National Laboratory to use machine learning to optimize the process of nuclear waste vitrification—essentially the safe containment and storage of nuclear waste. By improving the efficiency of this process, they could save years and billions of dollars in waste processing costs, a significant achievement given the high stakes of nuclear waste management.
John Hedengren, another chemical engineering professor involved in the research, underscores the importance of this work: “At $2 billion per year to operate the vitrification facility in Washington state, there is a big incentive to improve the efficiency.”
As the world grapples with the twin challenges of climate change and energy security, nuclear power is increasingly seen as a critical component of the clean energy mix. However, it still faces significant public skepticism due to concerns about safety, cost, and waste. Memmott hopes their research will help change that narrative.
“People are worried about the waste and the cost, but things like that are being solved,” Memmott asserts. “Shift the focus away from ‘Nuclear is the expensive, not desirable option’ to ‘With the right tools, we can do it really well.’”
The work being done at BYU is a testament to the power of innovation and the transformative potential of AI in the energy sector. As Memmott and his team continue to push the boundaries of what’s possible, they’re not just designing better nuclear reactors—they’re paving the way for a future where clean, safe, and reliable energy is within reach for all.
Comments