UK and US Link AI Supercomputers to Fast-Track Fusion Energy with Digital Twins

UKAEA PPPL Partnership Creates Transatlantic Computing Federation

The UK Atomic Energy Authority and Princeton Plasma Physics Laboratory revealed plans on September 14 at the Global Fusion Policy Summit in London to connect their dedicated AI supercomputers in an unprecedented cross-border scientific cooperation effort. The proposed SUNRISE-STELLAR-AI Federation would link UKAEA's £45 million SUNRISE platform with PPPL's $13 million STELLAR-AI system, creating a federated system for AI models that could reshape how researchers approach fusion energy development1

1

2

The partnership builds on a memorandum of understanding signed between UKAEA and PPPL in June and represents part of the U.S. Department of Energy's Genesis Mission to unite national labs, industry and academia. SUNRISE serves as the UK's first AI supercomputer dedicated to fusion energy, while STELLAR-AI provides AI and high-performance computing capabilities designed to host AI tools and workflows as they mature2

2

Source: Interesting Engineering

Training AI Models on Experimental Fusion Reactor Data

Researchers plan to use experimental fusion reactor data from MAST Upgrade and NSTX-U facilities—the Mega Amp Spherical Tokamak Upgrade in Oxfordshire and the National Spherical Torus Experiment-Upgrade in New Jersey. Both are spherical tokamak facilities with similar designs shaped like cored apples that use magnetic fields to confine plasma behavior1

1

2

The design similarity matters because machine learning systems trained on results from only one fusion machine often struggle when applied to another. By combining datasets from both facilities, researchers aim to build AI models that capture more of the underlying physics and make more reliable predictions across different machines. "Our goal is to let models and experiments move freely between the two systems. We will turn a collection of supercomputers into a single engine for fusion discovery," said Shantenu Jha, Head of Computational Sciences at PPPL1

1

Building Digital Twins for Fusion Reactors

The federation aims to support development of digital twins for fusion reactors at both laboratories. These virtual models would use data from experiments to create detailed simulations of fusion machines, allowing researchers to test changes and predict machine behavior in software before attempting them on physical equipment. The computers could fill gaps in experimental datasets through simulations, extending what researchers can study without running every scenario on actual hardware1

1

Rob Akers, Director of Computing Programmes at UKAEA, explained that the two laboratories could combine experimental data with simulations to explore operating conditions not yet tested. "Together, we can develop digital twins of both machines to support the design of future fusion power plants, creating models of spherical tokamaks that are more predictive, more actionable and ultimately more useful to fusion engineers,"1

1

Related Stories

US launches STELLAR-AI to slash nuclear fusion simulation time from months to real-time

UK government invests £45 million in AI supercomputer to accelerate fusion research

AI Breakthrough Accelerates Fusion Reactor Safety Calculations

AI System Controls Fusion Plasma in Milliseconds, Predicts Instabilities 200ms Before They Occur

Accelerating Path to Commercial Fusion Power Plants

The federation could eventually support work on future projects including the UK's Spherical Tokamak for Energy Production, known as STEP, and PPPL's proposed Spherical Tokamak Advanced Reactor. Jonathan Menard, chief scientist at PPPL, emphasized the collaborative necessity: "A fusion power plant is one of the most complex machines humanity has ever tried to build, and no single laboratory or nation will design it alone"2

2

Joe Milnes, Executive Director for Engineering and Computing at UKAEA, reinforced this view: "Fusion is one of the great scientific and engineering challenges of our time. To solve these challenges, fusion needs partnerships. The U.S. and UK are two global leaders in fusion research, and the federation of SUNRISE and STELLAR-AI can build on a long history of transatlantic cooperation to further advance fusion development"1

1

Researchers also plan to examine how computing jobs can move between the two platforms despite their different hardware, allowing scientists to select the system best suited to particular computational tasks. The partners aim to shorten the design cycle for future fusion systems, make better use of existing experiments and give the next generation of fusion scientists a shared platform. They plan to expand the federation over time, with the long-term goal of a shared foundation model for spherical tokamaks and eventually a wider range of tokamak configurations2

2

The proposed federation remains at the exploratory stage, with partners still working through technical requirements and next steps to make the operational shared supercomputer a reality1

1