by The digiLab Team
Updated 10 January 2024
Tokamak Energy collaborate with digiLab
digiLab hosted a Machine Learning (ML) workshop at Tokamak Energy, focusing on the application of mission-critical ML to fusion energy development workflows. Attendees discovered how to use digiLab’s ML platform, twinLab, to combine their fusion expertise with the latest in probabilistic modelling and uncertainty quantification (UQ).
Tokamak Energy has already proved itself as the pioneering fusion energy company, becoming the first private developer to reach the landmark plasma ion temperature of 100M °C. For the next stage of commercialisation, Tokamak Energy is focusing its efforts on developing the technology needed for fusion power plants. This means pushing the boundaries of HTS magnet design, material science, plasma physics, and control systems design.
To minimise the timescales of R&D pathways and the capital cost of the plant, models and predictions made today must not only define the performance of future facilities, but also provide assurance about the pathways chosen in regions of high uncertainty and limited data.
Tokamak Energy recognised that traditional simulations can be greatly accelerated allowing a much higher level of fidelity, and engaged digiLab, leaders in explainable AI for safety-critical applications, to help accelerate the adoption of ML in fusion and introduce the twinLab platform for building fast emulators with built-in uncertainty quantification.
digiLab has a five-year partnership with the UK Atomic Energy Authority, acting as Official Partner and Advisor in UQ, with the goal of introducing to the fusion ecosystem ML solutions ranging from advanced reactor control to systems engineering design. The fruits of this partnership are now enabling advancements in the private sector, where digiLab can combine the power of twinLab with solution engineering and skills training from a 30-person strong team of experts in ML, AI, data science, and product development.
The workshop was led by digiLab’s CTO and Turing AI Fellow, Professor Tim Dodwell, who said: “Utilising explainable ML in R&D both accelerates and de-risks the development of fusion energy. I believe Tokamak Energy is well-positioned to deliver on the commercialisation of fusion, and I look forward to assisting them with the implementation of novel ML techniques in their development pathway.”
Michael Porton, Tokamak Energy Chief Engineer, said: “We’re exploring a full range of technologies to support our mission to deliver clean, secure, affordable fusion power to the world, addressing the twin challenges of energy security and climate change. Machine learning can accelerate progress and we’re excited about digging deeper and unlocking its full potential.”
digiLab operates as the ML partner for industry, with a team of over 30 experts in ML, AI, data science, and software engineering. As the world increasingly adopts data-driven approaches, digiLab leads in areas where data is uncertain or sparse, where other techniques cannot perform. Taking a Bayesian view of the world, digiLab believes in the value of eliciting expert opinion from engineers and scientists, while capturing the inherent uncertainty in modelled systems to enable informed and optimum decision making. digiLab’s flagship offering is twinLab, the ML platform which enables engineers to build fast emulators with built-in uncertainty quantification, supported by our solution engineering team and ML skills training.
Get in-touch: email@example.com
About Tokamak Energy
Tokamak Energy is a leading global commercial fusion energy company based near Oxford, UK. We have an unrivalled track record designing and operating spherical tokamaks; the optimal route to commercial fusion energy.
In addition to fusion energy, Tokamak Energy is recognised as the leader in High Temperature Superconductor (HTS) magnet design, numerical modelling, and prototyping. Our dedicated HTS magnet team, in collaboration with key manufacturing partners, is focussed on becoming the leading supplier of HTS magnets to multiple markets.
The company, founded in 2009 as a spin-off from the UK Atomic Energy Authority, currently employs a growing team of over 250 people with experts from the UK and around the world. It combines world leading scientific, engineering, industrial and commercial capabilities. The company has 70 families of patent applications and has raised $ 250 million, comprising $ 200m from private investors and $ 50m from the UK and US governments. Our US subsidiary, Tokamak Energy Inc, was established in 2019. Visit www.tokamakenergy.co.uk for more information.
If you found this post helpful, you might enjoy some of these other news updates.
The Ethical AI Database
digiLab is part of the Ethical AI Database
Dr Andy Corbett
Understanding Uncertainty Quantification: The Different Types
Understanding Uncertainty Quantification - an article on the different types.
Prof Tim Dodwell
Tech Nation Rising Stars 5.0 - delighted national winners!
Tech Nation Rising Stars 5.0 winners
The digiLab Team