Machine learning techniques predict optimum fusion energy scenarios faster than ever before. Research from the DIII-D National Fusion Facility.
Research
Figuring out how tungsten moves through a fusion energy device is important in order to design reactors. Research at the DIII-D National Fusion Facility developed a new approach to modeling that can make it easier to get accurate predictions.
Improving fusion plasma codes and demonstrating the ability to accurately model plasma behavior in experiments. Research led by UC Irvine at the DIII-D National Fusion Facility.
Machine learning is used to identify plasma instabilities in fusion energy reactors. Ph.D. thesis research performed at the DIII-D National Fusion Facility.
Learn about all the ways the DIII-D National Fusion Facility will contribute to the development of fusion energy over the next many years. Part of a special collection in Physics of Plasmas, “Private Fusion Research: Opportunities and Challenges in Plasma Science.”
Graduate students at the DIII-D National Fusion Facility have built a large language model (LLM) from the entire archive of experiment journal notes. The resulting interface provides remarkably good answers to questions about how to improve fusion performance in experiments.