Keynote Speakers

Professor Lee Margetts is UKAEA Chair of Digital Engineering for Fusion Energy at The University of Manchester and Director of the Fusion Engineering Centre for Doctoral Training, helping to develop the next generation of researchers and engineering leaders for the fusion sector.

For centuries, engineers have built machines that are powerful, precise and transformative. Yet even the most advanced machines remain fundamentally passive. They do not truly understand their condition, anticipate their future, or learn from their experience. The next revolution in engineering is changing that.

Advances in imaging, sensing, simulation, high-performance computing and artificial intelligence are converging to create something entirely new: digital systems that can continuously observe, understand and predict the behaviour of complex physical assets. Technologies that once revealed hidden structures through X-rays and laser scanning are now part of a broader vision in which data is transformed into insight, and insight into action.

Fusion energy presents one of the greatest opportunities to realise this vision. Future fusion power plants will be among the most complex machines humanity has ever built. Their success will depend not only on breakthroughs in physics and materials science, but also on the creation of a digital nervous system: a living digital counterpart that continuously senses, learns and anticipates. The future of fusion is not simply about building a star on Earth. It is about giving that star the ability to understand itself.

Fusion is more than an energy challenge. It is a proving ground for a new generation of intelligent engineering systems capable of accelerating discovery, reducing uncertainty and transforming how complex technologies are designed, operated and maintained. If the twentieth century was defined by machines that amplified human strength, the twenty-first may be defined by machines that amplify human understanding.

This talk explores how the convergence of imaging, computation, digital twins and AI is reshaping engineering, and why fusion energy provides a unique opportunity to lead that transformation. The challenge now is not merely to build better reactors, but to create the digital intelligence that will enable them to evolve, adapt and improve. In doing so, we have the opportunity not only to accelerate the path to clean, abundant energy, but also to redefine the relationship between humans, machines and the engineered world.

Abstract: Yarn-path extraction in large 3D textiles based on periodicity and volume registration

Segmenting yarn paths in 3D woven reinforcements from low-resolution tomographic images is notoriously difficult and time-consuming, and is still performed manually today.   The chosen example is the segmentation of warp and weft yarns at the root of a Leap fan blade, with a voxel size of 140 µm, and the image extending over the entire part.

The initial observation is that, in some regions, the weaving topology is periodic.  From an autocorrelation analysis, a unit periodic cell is easily identified, and, by exploiting its periodicity, warp and weft yarns are segmented within the cell.  This task is accomplished using local matching with a template of the yarn cross-section, exploiting the continuity of each yarn path and the consistency of the warp and weft paths in 3D space, with no overlap. The unit cell is tiled along the three space directions to create a reference volume of arbitrary size.  Global Digital Volume Correlation registers the actual image with the periodic reference, accounting for large-scale distortions to match the two 3D images.  The global approach is based on a fine-mesh discretization of the displacement field and mechanical regularization.  These two features help avoiding local-minima trapping, as can be feared in an almost periodic medium. The mapping can now be used to transfer the unit-cell segmentation to the entire volume, thereby providing a sound labeling of each individual yarn.