The Institute for Plasmas and Nuclear Fusion (IPFN), Instituto Superior Técnico and Proxima Fusion, signed a cooperation agreement to commercialise fusion energy, based on stellarator technology, a class of devices that hold high potential for long-duration plasma confinement.
The work developed by the Técnico team involves sophisticated numerical methods and innovative algorithmic approaches, driving the design of optimized magnetic coil configurations and plasma geometries. The focus of this collaboration includes the exploration and application of fast optimization loops and machine-learning-inspired techniques – the cutting-edge tools that will likely shape the future of fusion energy.
The Técnico Professor Rogerio Jorge, who participates in the project, highlights “it’s a step towards a sustainable energy future”. The race to commercialise fusion energy is critical because as an energy source, fusion is both carbon-neutral and potentially nearly limitless, with fusion reactors using hydrogen isotopes and lithium as fuel.
Unlike conventional nuclear fission reactors, fusion power also avoids creating long-lived, high-level nuclear waste and the associated safety risks.
Lucio Milanese, co-founder and COO of Proxima Fusion, said: “By partnering with Técnico, we are tapping into a wealth of scientific knowledge and computational expertise. The integration of advanced computing techniques will enable us to iterate on designs more efficiently and accurately”.
Stellarators: harnessing the immense potential of fusion energy
Stellarators, unlike the more conventional tokamaks, have a unique edge: they maintain plasma confinement without the necessity of a plasma current. This crucial distinction means that fusion power plants built with stellarators can operate continuously, with significant operational and commercial benefits. However, the intricacies of stellarators, with their complex, non-symmetric magnetic field configurations, present substantial optimization challenges.
Computational optimization plays a critical role in refining these configurations to maximize plasma stability and minimize losses, two crucial factors to make fusion an effective energy source for humanity. Advanced optimization techniques enable rapid, iterative improvements to stellarator designs.
By swiftly evaluating and learning from thousands of potential coil geometries and magnetic field configurations, these techniques drive the development of stellarator designs that offer optimal performance. Optimization also aids in reducing engineering challenges associated with coil fabrication and maintenance, thus making stellarators more practical for commercial applications. The end result is a stellarator that brings us closer to harnessing the immense potential of fusion energy.
