US Department of Energy Supports Artificial Intelligence Project for Particle Accelerators

As part of its Genesis Mission to accelerate scientific discovery, the US Department of Energy has awarded second-phase funding to the MOAT-Core project, which uses artificial intelligence to optimize the design and operational processes of particle accelerators.

Managed by Lawrence Berkeley National Laboratory, the Multi-Office Accelerator Team Core (MOAT-Core) project aims to increase the efficiency of particle accelerators used in physics, chemistry, biology, and materials science. At the heart of the project is 'Osprey', an AI assistant initially used at Berkeley Lab's Advanced Light Source facility, which has now been transformed into a multi-facility platform.

With the new funding, the project has been expanded to 16 institutions, including Argonne, Brookhaven, Fermi, Thomas Jefferson, Oak Ridge, Pacific Northwest, and SLAC national laboratories, as well as various universities. Osprey allows operators to set up experiments, precisely tune particle beams, and troubleshoot equipment issues using natural language commands.

In future phases, it is planned to integrate this technology with digital twins to be used in next-generation accelerator designs, potentially shortening computational timelines by years. This development is expected to contribute to research in fields such as microelectronics, energy storage, and quantum computing.