In pursuit of autonomous nuclear plant operations
It seems the analysis for the grasp’s was simply the tip of the iceberg. There was nonetheless work left to be performed.
For the long run viability of nuclear energy, small crops, positioned in rural areas, are a definite risk. To this point, operations performed in legacy crops have concerned intensive handbook operations, that are viable as a result of these amenities function at one hundred pc capability and the ability delivered can justify the prices of sustaining a big employees. However microreactors distributed at scale and in distant areas can’t afford a big bench of human expertise. It’s the place supervised and totally vetted autonomous operations will assist.
A main query was: “How will we transition to autonomous operations in nuclear energy crops?” Fortier wished one built-in strategy, a central supervisory management system as a substitute of many interlinked components. Whereas a reliable requirement, the problem right here was in modifying operations that had been primarily meant for people to now additionally embody machines.
Fortier realized that regardless of how thorough the framework for a supervisory management system, its scope could be severely restricted if its rigidity couldn’t accommodate each people and machines. “As a result of every thing is human-centric, it doesn’t can help you select one of the simplest ways to do a process,” Fortier observes. What if, as a substitute, people and computer systems may tag-team and do what every does greatest, with strategic human intervention solely delivered as obligatory?
MIT-facilitated collaborations
The aim was elegant, however its scope prolonged far past a grasp’s thesis. A doctorate appeared like a pure development, so Fortier has continued her analysis towards a PhD after finishing her grasp’s in 2025.
It was in pursuit of autonomous operations that Fortier found the ability of collaborations at MIT. Her analysis advisor, Sacit Cetiner, has a joint appointment with MIT NSE and the Idaho Nationwide Laboratory (INL). To handle the problem of devising an autonomous supervisory management system with an efficient and simply adopted human-machine interface, Fortier labored with Katya Le Blanc, a senior human components scientist at INL. A collaboration with the Human System Simulation Laboratory at INL helped Fortier develop a greater understanding of designing cyber-physical programs.
“I’m very a lot an engineer and don’t have a variety of expertise in human habits, so the collaboration with INL was an enormous profit for me. I obtained higher insights into many facets, together with what you need to see when a human has to take over for a machine when it’s now not working,” Fortier says.
She additionally availed of collaborations with Westinghouse, a number one design group and vendor for current- and next-generation nuclear energy crops — Fortier accomplished a summer season internship there in 2025 — to check drive concepts she had about autonomous operations options.
At MIT, Fortier realized management concept from one in every of her co-advisors, Anuradha Annaswamy, a founder and director of the Lively-Adaptive Management Laboratory within the Division of Mechanical Engineering. “She’s a management programs professional, which actually advantages me as a result of whereas I can clarify what to do with a nuclear energy plant, she can assist me perceive higher easy methods to go about operations from a management programs perspective,” Fortier says.
Annaswamy advises Fortier on supervisory management system framework and execution. Fortier has additionally been taking lessons associated to regulate programs associated lessons to construct a basis within the self-discipline. Curtis Smith, the previous director for INL’s Nuclear Security and Regulatory Analysis Division, and now KEPCO Professor of the Apply of Nuclear Science and Engineering at MIT NSE, is Fortier’s different co-advisor.
A step-by-step development towards autonomy
Fortier clarifies that the operations programs she’s designing will incorporate a gradual and systematic transfer towards autonomy in order to construct belief with customers. “After we introduce an automatic process that walks you step-by-step by way of what you’d be doing anyway, it’s reassuring and builds belief,” Fortier factors out. Her doctoral work focuses on incorporating objective-oriented operations the place a management system can create the sequence of occasions wanted to achieve the target, fairly than observe a predetermined working process.
One other level of reassurance is that Fortier is creating automation based mostly on a course of known as finite state automata, which, in contrast to AI, could be very clear in its execution. Some great benefits of finite state automata, which she additionally studied extensively throughout an internship at INL in summer season 2024, is that it’s a discrete occasion system. Because of this each transfer within the automation framework is event-driven — if this occurs, do this — so it adjusts for present situations within the plant and transitions between numerous states or occasions very clearly. It’s addressing a posh drawback by way of standard automation, not AI-driven automation. “We’re not utilizing a data-driven statistical strategy like machine studying as a result of we don’t but have the instruments to validate the operation of such programs,” Fortier says.
Future affect
So promising is Fortier’s work on creating automation for nuclear crops that she was one of many winners of the 2025 version of the Improvements in Nuclear Power Analysis and Improvement Scholar Competitors from the Division of Power’s Nuclear Power College Program.
Utilizing the nuclear plant automation program on next-generation tools will ship obligatory traction in creating and deploying industrial microreactors.
However first will come the work of scaling the supervisory management system, borrowing from insights gained from work on a small side of management. Fortier is happy concerning the street and the probabilities that await. “The collaborations with different individuals, and the relationships we’ve established with stakeholders, have actually helped make an affect and supported the relevancy of the work,” she says. “Typically while you’re caught in your personal bubble, that exterior perspective is de facto helpful.”
