The AI: Known as HEAT-ML, the new AI could lay the foundation for software that speeds up the design of future fusion systems. According to the article, such software could also enable good decision-making during fusion operations by adjusting the plasma so that potential problems are thwarted before they start.
“This research shows that you can take an existing code and create an AI surrogate that will speed up your ability to get useful answers, and it opens up interesting avenues in terms of control and scenario planning,” said Michael Churchill, head of digital engineering at PPPL and coauthor of a paper in Fusion Engineering and Design about HEAT-ML.
A challenge: To harness fusion, researchers need to overcome scientific and engineering challenges, including handling the intense heat coming from the plasma, which reaches temperatures hotter than the sun’s core when confined using magnetic fields in the tokamak fusion vessel. AI can speed up the calculations that predict where the heat will hit in the tokamak.
“The plasma-facing components of the tokamak might come in contact with the plasma, which is very hot and can melt or damage these elements,” said Doménica Corona Rivera, an associate research physicist at PPPL. “The worst thing that can happen is that you would have to stop operations.”
In focus: HEAT-ML was specifically made to simulate a small part of SPARC, the tokamak currently under construction by CFS. The company hopes to demonstrate net energy gain by 2027.
The research team has focused on a section of SPARC where the most intense plasma heat exhaust intersects with the material wall. The section is made up of 15 tiles near the bottom of the machine and is subjected to the most heat.
To create a simulation, the researchers generated what they call shadow masks—3D maps of magnetic shadows. These shadows are areas on the surfaces of a fusion system’s internal components that are shielded from direct heat.
Creating simulations: HEAT-ML traces magnetic field lines from the surface of a component to see if the line intersects other internal parts of the tokamak. The article explains that if an intersection occurs, the region is marked as shadowed. “However, tracing these lines and finding where they intersect the detailed 3D machine geometry was a significant bottleneck in the process. It could take around 30 minutes for a single simulation and even longer for some complex geometries,” the article noted.
HEAT-ML accelerates the calculations to a few milliseconds. Its neural network was trained using a database of about 1,000 SPARC simulations to learn how to calculate shadow masks.
HEAT-ML currently works only on the specific design of SPARC’s exhaust system, but the research team is looking to expand its capabilities for calculation of shadow masks for exhaust systems of any shape and size, as well as the rest of the plasma-facing components inside a tokamak.
