Part 1 of 2 Parts
     Although nuclear energy currently provides more low-carbon electricity in the U.S. than solar power and wind power put together, the U.S. fleet of commercial nuclear power reactors is aging. U.S. commercial nuclear power plants are becoming ever more expensive to maintain while the cost of alternative renewable energy is steadily dropping. Commercial nuclear power is struggling to survive in an increasingly competitive energy marketplace.
     One of the possible places that the cost of generating nuclear power can be reduced is inside the reactor core where the fission process takes place. The exact placement of the nuclear fuel rods which contain pellets of nuclear fuel has a significant impact on the efficiency of the fission reaction. If the rods can be placed in optimal positions, they burn less fuel and require less maintenance. Nuclear engineers have spent decades of trial and error to design better layouts that will extend the life of expensive fuel rods. Now researchers are working on ways to use artificial intelligence to improve fuel rod layouts.
     Researchers at the Massachusetts Institute of Technology and Exelon have found that by turning the design process into something resembling a computer game, an AI system can be trained to generate and explore dozens of possible optimal configurations that can increase the average lifespan of each fuel rod about five. This could save the owners of a typical commercial nuclear power plant as much as three million dollars a year.  The AI system being studied can also explore and identify optimal solutions much more quickly than any human being. Designs can be generated and modified rapidly in a safe, simulated environment. The MIT and Exelon team published the results of their research in the journal Nuclear Engineering and Design this month.
     Koroush Shirvan is an assistant professor in MIT’s Department of Nuclear Science and Engineering and the senior author of the AI study.  He said, “This technology can be applied to any nuclear reactor in the world. By improving the economics of nuclear energy, which supplies 20 percent of the electricity generated in the U.S., we can help limit the growth of global carbon emissions and attract the best young talents to this important clean-energy sector.”
     In the core of a typical commercial nuclear power reactor, the fuel rods are arranged in a grid according to the levels of uranium and gadolinium oxide they contain. The radioactive uranium in the fuel rods provides the fission reactions that generate energy and the rare-earth gadolinium slows down the fission reactions to maintain a safe steady operation. In an ideal arrangement, these competing processes balance each other in order to provide the most efficient utilization of the fuel in the rods. Human designed rod arrangements have been subjected to traditional algorithms to improve their layout. However, in a standard one hundred rod assembly, there are a huge number of options that must be evaluated. This approach to optimization has had limited success to date.
Please read Part 2 next