Thesis: Accounting for Uncertainties in the Design of Future High-Field Tokamaks

Proposed Thesis at the IRFM (CEA Cadarache)
Supervisors: Eric Nardon ([email protected]), Yanick Sarazin ([email protected])
Desired start date: late 2026

Summary of the topic
The most promising approach to harnessing nuclear fusion energy for civilian purposes uses intense magnetic field
to confine a hot plasma and limit its interaction with the solid walls
of the chamber. This is referred to as “magnetic fusion.” The most advanced concept in this field is
the tokamak. The advent on an industrial scale of high-temperature superconductors
(HTS), capable of generating magnetic fields twice as strong as the superconductors
used in this field until now, could be highly beneficial, but the matter remains undecided. This
topic is the focus of the SF-Plant project coordinated by the IRFM within the PEPR research program
SupraFusion, under which this thesis is being conducted.
The proposed thesis aims to address the issue of uncertainties in the design of power-generating tokamaks,
particularly those using HTS. The models used for this design work have limited
accuracy due to the complexity of fusion plasma physics and tokamak engineering.
The initial goal will be to quantify the effect of these uncertainties on the robustness and expected performance
of a given design (i.e., for fixed parameters—geometry, magnetic field
, etc.). In a second phase, the goal will be to develop and implement a design optimization method
that inherently accounts for uncertainties. This work may draw upon
existing simulation codes such as D0FUS [1] and SYCOMORE [2], which are key tools
for the design of future tokamaks. Reference [3] provides a foundation from which we can draw inspiration.

References
[1] T. Auclair et al., “The tokamak system code D0FUS and its first applications: Impact of B- and
confinement scaling laws on power plant design,” https://doi.org/10.1016/j.fusengdes.2025.115270
[2] S. Kahn et al., “Sensitivity analysis of fusion power plant designs using the SYCOMORE system-
-code,” Nuclear Fusion 2020, https://doi.org/10.1088/1741-4326/ab4879
[3] M. Coleman et al., “Definition of an EU-DEMO design point robust to epistemic plasma physics
uncertainties,” Nuclear Fusion 2025, https://doi.org/10.1088/1741-4326/ada6d9