University of Cambridge > Talks.cam > Data Intensive Science Seminar Series > Hot & cool: tackling the Fusion grand challenge with AI and HPC

Hot & cool: tackling the Fusion grand challenge with AI and HPC

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact James Fergusson.

Fusion holds the promise to deliver carbon free, almost limitless amounts of energy to power our world. A promising avenue to achieve Fusion is the Tokamak, a doughnut-shaped machine where plasma, a quasi-neutral ionised gas, is confined by means of magnetic fields. The UK has the ambition to deliver the Spherical Tokamak for Energy Production (STEP) by 2040 – this will be a fully working Fusion power plant. A number hurdles still need to be overcome on the way to making Fusion a reality. However, we now have access to some powerful tools unavailable to the scientists and engineers of the past: modern AI and exascale HPC solutions may just finally unlock commercial Fusion power. In this talk I will introduce some of the challenges that are unique to Fusion, and will showcase how these are approached with a holistic view at the UK Atomic Energy Authority. I will then take a dive into our efforts within the Advanced Computing Department to tame Fusion using HPC and AI.

Lorenzo is an AI Specialist in the Advanced Computing Department at the UK Atomic Energy Authority researching AI applications for Fusion energy. Prior to joining UKAEA in October 2021 he obtained a PhD in Astrophysics at the University of Southampton where he used deep generative models to compare galaxies in simulations to real galaxies. He also devised Monte Carlo methods to study galaxy structure and he applied similar techniques to modelling blood pressure treatment in presence of measurement uncertainty, for which he won an early-career STEM for Britain silver medal awarded at Westminster. He is interested in any and all AI methods that can speed up the delivery of Fusion power and believes in collaborative science as the only means to achieve big scientific breakthroughs. To this end, he recently won an award by the Alan Turing Institute to promote Fusion as a use case in the AI and Computer Science community. You can find more details about my Turing grant at https://www.gov.uk/government/news/ai-has-star-power-to-accelerate-fusion

This talk is part of the Data Intensive Science Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2022 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity