The U.S. Department of Energy (DoE) announced new funding for a project led by researchers at MIT’s Plasma Science and Fusion Center (PSFC) and four partner institutions as part of a strategy to accelerate the arrival of fusion energy and achieve carbon neutrality by 2050.
New work being carried out by MIT's Cristina Rea and her co-researchers aims to accelerate progress in fusion science and make fusion energy a reality as soon as possible. Image source: AnaKova/U.S. Fusion Outreach Department
As the effects of climate change continue to grow, so does interest in the potential of nuclear fusion as a clean energy source. Although fusion reactions have been studied in laboratories since the 1930s, there are many key questions that scientists must answer before fusion power generation becomes a reality, and time is of the essence.
Enhance research capabilities through data and artificial intelligence
Cristina Rea, a research scientist and group leader at the Center for Plasma Science and Fusion, will serve as principal investigator on a newly funded three-year collaborative project to pilot integrating fusion data into a system that can be read by artificial intelligence tools. PSFC, along with scientists from the College of William and Mary, the University of Wisconsin-Madison, Auburn University and the nonprofit HDF Group, plans to create a holistic converged data platform, elements of which could provide unprecedented access to researchers, especially underrepresented students. The project aims to encourage diverse groups in academia and the workforce to engage in convergence and data science through an outreach program led by the group’s co-investigators, four out of five of whom are women.
Cristina Rea (right) is a principal investigator on a new U.S. Department of Energy-funded project that will seek to modernize data from fusion devices such as Alcatel's C-Mod tokamak (background) to harness the power of artificial intelligence. Danas Ridikas, head of the International Atomic Energy Agency's physics department, wrote a letter of support for Rhea's proposal. Photo credit: EarlMarmar
The DOE award, part of a $29 million package of funding for seven projects at 19 agencies, will support the group's distribution of data generated by fusion devices such as PSFC's Alcator C-Mod, a donut-shaped "tokamak" that uses powerful magnets to control and confine fusion reactions. Alcator C-Mod ran from 1991 to 2016, and its data are still being studied, thanks in part to the PSFC's commitment to the free exchange of knowledge.
Addressing data acquisition challenges
Currently, there are nearly 50 public magnetic confinement fusion experimental devices; however, historical and current data from these devices are difficult to obtain. Some fusion databases require a signed user agreement, and not all data is cataloged and organized in the same way. In addition, it is difficult to use machine learning (a type of artificial intelligence tool) to analyze data and achieve scientific discoveries without time-consuming data reorganization. As a result, fewer scientists are working on convergence research, barriers to discovery have increased, and bottlenecks are encountered in using AI to accelerate progress.
The project’s proposed data platform addresses technical barriers to greater transparency and inclusivity in science through FAIR (Findable, Interoperable, Accessible, Reusable) and UNESCO’s Open Science (OS) recommendations; as required by the Ministry of Education, all deliverables by researchers will adhere to FAIR and OS principles.
The platform's database will be built using MDSplusML, an upgraded version of the MDSplus open source software developed by PSFC researchers in the 1980s to catalog experimental results from Alcator C-Mod. Today, nearly 40 fusion research institutions use MDSplus to store and provide fusion data to the outside world. The release of MDSplusML aims to continue this tradition of open collaboration.
Promoting diversity and inclusion in fusion research
The researchers intend to address barriers to women and underrepresented groups participating in research, not only by improving general access to fusion data, but also through a subsidized summer school that will be held at William & Mary over the next three years and focus on topics at the intersection of fusion and machine learning.
Regarding the importance of their research, Rhea said: "This project is about filling a need in the fusion field and laying the foundation for our success. Scientific progress in the fusion field is achieved through multidisciplinary collaboration and cross-fertilization, so accessibility is an absolute necessity. I think we all understand now that diverse communities have more diverse ideas, and they are able to solve problems faster."
The collaborative project's work is also consistent with important research areas identified by the International Atomic Energy Agency's "Fusion Artificial Intelligence" Coordinated Research Program (CRP). Rea was selected as technical coordinator for the International Atomic Energy Agency's CRP, which emphasizes community engagement and knowledge acquisition to accelerate nuclear fusion research and development. In a letter supporting the team's proposed project, the IAEA noted: "The work [researchers] will perform [...] will benefit not only our CRP, but also the international fusion community as a whole."
Dennis Whyte, director of the PSFC and professor of engineering at Hitachi America, added, "I'm pleased to see the PSFC and our collaborators at the forefront of applying new artificial intelligence tools, while encouraging and enabling the extraction of critical data from our experiments."
"Having the opportunity to lead such an important project is profound, and I feel a responsibility to demonstrate that women are leaders in STEM fields," Rhea said. "We have an incredible team of people who are deeply motivated to improve our fusion ecosystem and contribute to the realization of fusion energy."