The experiments on DIIID showed that the AI model could forecast tearing mode instabilities up to 300 milliseconds in advance, enough time for an AI controller to adjust the operating parameters of the experiment to avoid the instabilities.
The team believes the findings show that dynamic control of a fusion plasma using AI could solve this and other plasma instabilities that stand in the way of sustained operation of future fusion power plants.
The research was published in Nature magazine on February 22 and is described in a PPPL press release: