In a study led by the Department of Energy's Oak Ridge National Laboratory, researchers have developed an AI model that helps identify new alloys to be used as shielding in nuclear fusion facilities.
The study, published in Scientific Data, is a major step forward in improving nuclear fusion facilities. Traditionally, alloys used in fusion facilities were made primarily of tungsten. While these alloys did prove resistant against high temperatures, they were inconsistent in maintaining shielding.
"Recently, the material science community has explored the opportunity of replacing these standard technology materials with something completely new and disruptive," said ORNL AI data scientist Massimiliano Lupo Pasini.
Identifying novel metallic combinations proved challenging when considering the vast number of possible combinations. To bypass this seemingly endless trial and error period, the team developed an AI model to identify alloy candidates more efficiently.
"We are trying to help the material scientists with their trial-and-error approaches in identifying the relative percentage of the different elements that need to be mixed together in order to come up with alloys that can lead to disruptive technological advances in fusion," Lupo Pasini added.
To further optimize the model, the researchers intend to use data generated by the Perlmutter and Summit supercomputers to train the AI model.