Mihai Diaconeasa
ASSISTANT PROFESSOR OF NUCLEAR ENGINEERING AT NORTH CAROLINA STATE UNIVERSITY
BORN IN ROMANIA, MIHAI DIACONEASA CONSIDERS himself “pretty lucky.” Reflecting on his upbringing during the tumultuous post-communist years of the 1990s, he said, “With not a lot of perspective in life but still being dedicated and focusing on the things in front of me, and leveraging all the opportunities and benefiting from the mentoring I had along the way—I ended up here.”
As a young student with a passion for mathematics, Diaconeasa was intrigued by nuclear physics. But when he mentioned his interest to his middle school teacher, he was given a warning: “Don’t touch that. It’s too radioactive. If you want to live a long life, I wouldn’t go down that road.” While he understood his teacher’s concern, Diaconeasa said, “That’s like saying don’t change a light bulb because you’re going to get electrocuted.” So, despite the dissuasion, Diaconeasa was drawn to the discipline’s potential to address energy needs safely and sustainably.
Today, as an assistant professor and associate department head in the department of nuclear engineering at North Carolina State University, Diaconeasa names his proudest scientific contribution as his work on mathematically modeling how complex systems evolve in time. “We’re still pushing in that direction nowadays,” he said, “aided by reinforcement learning approaches to address the completeness problem of nuclear safety cases.”
His research focused on analytically developing and computationally solving a discretized form of the master equation describing the time evolution of nuclear reactors with human control under any conditions as being in probabilistic combination of states at any given time, he shared, starting from two of its specialized forms—the Fokker-Planck and Chapman-Kolmogorov equations.
Mihai Diaconeasa has a passion for tennis—not just for the competition, but for its mental focus, coordination, and underlying science. He sees the game through the lens of geometry and mechanics: The spin, trajectory, and motion of the ball all come down to mathematics and physics.
ACCELERATING REAL-WORLD IMPACT
A key initiative of his is OpenPRA.org, a nonprofit, open-source project designed to democratize access to probabilistic risk assessment tools.
“Everything we do in the nuclear community is pretty much closed,” he explained. “But those reasons, in my view, can be mitigated.” OpenPRA.org aims to bridge academia and commercial applications, especially for countries with limited resources. “With an open-source community effort, you can actually try to fill some of those gaps,” he said.
To accelerate real-world impact, Diaconeasa also co-founded True Clean Energy Technologies, a startup helping streamline the licensing of next-generation reactors. “Licensing of reactors is very expensive… and this technological gap can be partly addressed with software solutions, including probabilistic AI,” he said, likening the regulatory engagement approach to using tax-filing software instead of filling out PDFs by hand.
Looking ahead, he hopes to advance a new research project using GPUs and TPUs to solve large Boolean functions using Monte Carlo approaches and supporting the expansion of Boolean differential calculus theory—essential to risk analysis, cryptography, and AI.
Ultimately, Diaconeasa believes that scientists have a duty to build public trust through risk communication when it comes to misconceptions around nuclear. “The burden is on us, the ones that actually work with it every day, just as the burden would be on a medical doctor to explain to their patients why it’s important to have an X-ray or undergo radiation therapy.”
© 2025 The American Society of Mechanical Engineers. All rights reserved.