[Tech] Development of a new topological optimization model to improve the efficiency of components for nuclear fusion.


Nuclear fusion is one of the few technologies with the potential to meet growing energy demand using sustainable, safe and low emission energy sources. Inspired by these scientific and technological challenges, the Catalan community of organisations working in nuclear fusion technology is joining forces in the FusionCAT research project.

In this project, Eurecat is taking part in Project 3: Fusion reactor studies with its Advanced Simulation Line, part of its Product Innovation and Development Unit. More specifically, Eurecat is involved in Task 3.2: Modelling, analysing and designing accelerator components for fusion materials research facilities, leading Subtask 3: Thermo-mechanical and computational fluid dynamics studies of components exposed to the IFMIF-DONES accelerator beam and Subtask 4: Analysis of industrial partners to secure potential future contracts and collaborations as part of international fusion programmes.

Eurecat’s Advanced Simulation Line is made up of a team of expert engineers in numerical analysis with extensive experience in projects geared towards analysis, design and optimisation of products, structures, fluid-dynamic systems and processes. This involves using advanced design and simulation methods at a multi-physical level, such as the finite element method (FEM), computational fluid dynamics (CFD), generative design and topological optimisation.

This line is equipped with a number of simulation platforms including Ansys, Simulia and Altair. They deliver the capacity to numerically reproduce a wide range of physics in the mechanical, fluid-dynamic, rheological, heat transfer, electromagnetic and other fields.

Eurecat and the IREC have been working together for years on projects to analyse and design components exposed to the IFMIF-DONES accelerator beam. The high energy density of this accelerator’s beam means that these components work in critical conditions, especially in terms of temperature and thermal stress. This poses a major challenge as it can jeopardise the proper operation and safety of the entire system.

This challenge can be addressed mainly from two angles. Firstly, by using special materials which can safely withstand the temperatures and thermal stresses involved and also feature high barrier capacity against the accelerator’s beam. Eurecat and the IREC have been evaluating a number of materials options. However, although there are materials capable of withstanding the working conditions, they do not have a sufficient safety margin and/or present other important limitations, especially in terms of cost, availability in the sizes required and the complexity of machining and welding.

The other way in which this challenge can be tackled is through design, in particular with respect to the cooling circuit. Eurecat has worked on design and analysis of the components and relevant cooling circuits by developing multi-physical simulation models drawing on computational fluid dynamics (CFD) and the finite element method (FEM). They can numerically reproduce the turbulent regime of the fluid flow and the heat transfer throughout the system along with the resulting thermal stresses and distortion. Eurecat is using these models to run an iterative design-simulation optimisation process which fine-tunes the design of the component in line with the simulation results derived in each iteration until a design considered optimal is achieved.

This is a conventional optimisation method with significant limitations. Firstly, there is a considerable subjective aspect to it in which the engineer’s experience and ability to interpret results play a critical role. Secondly, it is a manual process which calls for a lot of work in each iteration, and this means that the number of iterations which can be performed is often restricted.

To address these limitations, in the FusionCAT project Eurecat is to develop a new topological optimisation method. Topological optimisation is a specific generative design method which is based on numerical simulation methods. It makes it possible to get the optimum geometry from a design space in line with specified objectives and constraints. This method is widely used in the structural field but not in fluid dynamics and heat transfer. The new topological optimisation method developed by Eurecat will be able to integrate these physics in such a way as to optimise the components exposed to the IFMIF-DONES accelerator beam at a hitherto unprecedented level.

It is also expected that the new method will enable optimisation to be conducted at a multi-scale level. This means that in addition to making it possible to decide which areas of the design space should have material or not, it will also be able to identify whether it is better for there to be material with a certain porosity in any particular area (for example, the area where the beam hits) and to ascertain how and what this porosity should be in order to attain the best results. Depending on the type of porosity set by the model, structures with ad-hoc porosity will be designed later on (for instance based on lattice structures or minimal surfaces) by means of generative design methods which can then be manufactured using 3D printing technologies.