The project will use the latest in artificial intelligence (AI) and computational tools to accelerate fusion power plant design, testing and optimisation. By creating a virtual model of the plant, Thea Energy aims to evaluate performance, refine designs and test operating scenarios before construction begins, reducing development time and costs.
The initiative supports the US Department of Energy’s Genesis Mission, which seeks to use AI to advance fusion energy and other scientific technologies.
Under the collaboration, NVIDIA will use its Omniverse platform and AI infrastructure to integrate models, software and operational data into a digital twin capable of analysing power plant performance at scale. Synopsys will contribute simulation expertise and AI-driven multiphysics modelling to assess key systems, including the breeding blanket, which converts fusion energy and protects magnetic components.
Argonne National Laboratory will provide expertise in neutronics analysis, blanket design and energy conversion systems, while PPPL will contribute plasma modelling capabilities and high-fidelity computational tools for simulating plasma behaviour under power-plant operating conditions.
