Game-playing AI technique may lead to cheaper nuclear energy

In this AI-designed layout for a boiling water reactor, fuel rods are ideally positioned around two fixed water rods to burn more efficiently. MIT researchers ran the equivalent of 36,000 simulations to find the optimal configurations. Colors correspond to varying amounts of uranium and gadolinium oxide in each rod. Image: Majdi Radaideh/MIT
Researchers at the Massachusetts Institute of Technology and Exelon show that by turning the nuclear fuel assembly design process into a game, an artificial intelligence system can be trained to generate dozens of optimal configurations that can make each fuel rod last about 5 percent longer, saving a typical power plant an estimated $3 million a year, the researchers report.
The AI system can also find optimal solutions faster than a human and can quickly modify designs in a safe, simulated environment. The results appear in the journal Nuclear Engineering and Design.

In 1970, a bit more than 50 years ago, then–ANS President Nunzio Palladino gave an evening lecture in Brussels to the newly formed Belgian local section of the American Nuclear Society. It was the seed for what would become the Belgian Nuclear Society, but the story starts even earlier than that.



Just released by a group called 18 for 0, the 47-page preliminary study 
The Department of Energy has begun the environmental review of its proposed Versatile Test Reactor (VTR), releasing a
Declaring small modular reactors to be “the next innovation that will help us reach net-zero emissions by 2050,” Canadian Minister of Natural Resources Seamus O’Regan last week introduced his government’s 