Sign In

Communications of the ACM

ACM Careers

Engineers Use AI to Wrangle Fusion Power for the Grid

View as: Print Mobile App Share:
Azarakhsh Jalalvand, Egemen Kolemen, and Ricardo Shousha at Princeton University's Andlinger Center

Princeton's Azarakhsh Jalalvand, Egemen Kolemen, and Ricardo Shousha are among the authors of the study published in Nature.

Credit: Adena Stevens / Princeton University

A team of engineers, physicists, and data scientists from Princeton University, the Princeton Plasma Physics Laboratory (PPPL), and Chung-Ang University in South Korea have harnessed the power of artificial intelligence to predict — and then avoid — the formation of a specific superheated plasma problem in real time.

In experiments at the DIII-D National Fusion Facility in San Diego, the researchers demonstrated that their model, trained only on past experimental data, could forecast potential plasma instabilities known as tearing mode instabilities up to 300 milliseconds in advance. Turns out that was plenty of time for the AI controller to change certain operating parameters to avoid what would have developed into a tear within the plasma's magnetic field lines, upsetting its equilibrium and opening the door for a reaction-ending escape.

"The AI could develop a final control policy that supported a stable, high-powered plasma regime in real time, at a real reactor," says research leader Egemen Kolemen, associate professor at Princeton's Andlinger Center for Energy and the Environment, and a staff research physicist at PPPL.

The research opens the door for more dynamic control of a fusion reaction than current approaches, and provides a foundation for using artificial intelligence to solve a broad range of plasma instabilities. The team's findings are published in the journal Nature.

From Princeton University
View Full Article


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account