Enhancing Stellarator Accessibility through Port Size Optimization

A. Baillod, E. J. Paul, T. Elder, J. M. Halpern · 2025

Read Enhancing Stellarator Accessibility through Port Size Optimization by A. Baillod, E. J. Paul, T. Elder, J. M. Halpern (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-12319.txt.

Introduction

Baillod and co-authors make room for bigger access ports in stellarator fusion devices without wrecking the magnetic field, in a 2025 paper. A stellarator's twisted coils leave little space to reach the plasma for repairs and instruments. The authors treat each port as a closed curve on the plasma boundary and fold port size directly into the coil optimization, a first for the field.

They chart the trade-off between field quality and port size as a Pareto front, and explain the best port placement with a current potential argument. Extra windowpane coils can push past that front entirely, giving both a good field and generous access. The work targets practical reactor maintenance.

Catalog record

CREATORA. Baillod, E. J. Paul, T. Elder, J. M. Halpern
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-12319

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Enhancing Stellarator Accessibility through Port Size Optimization A. Baillod, E. J. Paul, T. Elder, J. M. Halpern Abstract Access to the plasma chamber in a stellarator reactor is essential for maintenance and diagnostics. However, the complex geometry of stellarator coils, often characterized by their strong twisting, can severely limit the space available for access ports. This study introduces a novel optimization approach in which access ports are represented as closed curves on the plasma boundary. By carefully selecting a set of objectives and penalties related to the access port, we demonstrate the first stellarator coil optimization explicitly targeting improved access port size. The trade-off between magnetic field quality and port size is analyzed through the Pareto front of their respective objectives. The optimal location of a port is explained using a current potential approach. Finally, we show that additional shaping coils, such as windowpane coils, can enable the crossing of the Pareto front to achieve superior configurations. 1 Introduction Stellarators are magnetic fusion devices that confine a plasma with a magnetic field generated by external coils. In a stellarator reactor, the extreme heat and neutron fluxes generated by the fusion process, progressively damage the plasma facing components and the reactor’s blanket elements [ Knaster et al., 2016 ] . The strategy used to replace the damaged elements, called maintenance scheme (MS), is thus a crucial element of reactor design [ Boozer, 2021 , Boozer, 2024 ] . The MS has to be reliable, as multiple maintenance periods will be required in the lifetime of a power plant. For example, it was estimated that about 11 to 13 replacements of the first wall would be required in the lifetime of the ARIES-CS power plant [ El-Guebaly et al., 2008 ] . In addition, the MS has to be fast, as a longer maintenance period makes the power plant less financially attractive — a study by Ward et al. [2005] , where the PROCESS code [ Kovari et al., 2014 , Kovari et al., 2016 ] was applied to study a tokamak fusion power plant, showed that the cost of electricity would scale as \text{coe}\sim A^{-0.6} , where A is defined as the fraction of time where the power plant is operational. While this study focused on a tokamak power plant, it still shows an important correlation between the fusion power plant availability, and its cost of electricity. Usually, power plant availability is assumed to be of the order of 85\% [ El-Guebaly et al., 2008 ] . In addition to impacting the cost of electricity produced by a fusion power plant, its accessibility might also have consequences on the tritium economy — as identified by Abdou et al. [2021] , tritium production is paused during maintenance, but the amount of tritium in the blanket decreases by radioactive decay. Different MSs for stellarators have been discussed in the literature. Recently, Warmer et al. [2024] compared qualitatively four different schemes. The first one considers only the use of vertical access ports, as for the DEMO tokamak [ Bachmann et al., 2022 ] . The second relies on both vertical and horizontal access ports, while the third considers large vertical access ports, accessible by translating one or more large coils around the device. Finally, the last considered remote MS assumes splitting the stellarator in different sections that can be moved radially. The qualitative comparison by Warmer et al. [2024] lead to no clear winner, and further studies on each scheme viability were deemed necessary. It was however recognized that, in general, remote maintenance in a stellarator is expected to be more challenging than in a tokamak, as stellarators typically have more coils, their shaping can reduce port sizes, and the plasma chamber shaping complicates the replacement of internal components. The freedom in three-dimensional shaping of both the plasma and the coils can however also be an advantage, as one could promote accessibility in a stellarator optimization, as it is shown in this paper in the context of a port-based MS. Other approaches, such as segment-based MSs, are not discussed. Note that accessibility also depends on coil topologies, as discussed by Ku and Boozer [2010] . In this paper, we only consider modular and windowpane coils, and do not consider helical coils [ Yamaguchi et al., 2021 ] . It is worth mentioning here that large access ports are not only important for reactor accessibility, but also for experimental stellarators. Often, heating systems such as neutral beam heating (NBI), or diagnostics, require large access ports to the plasma. For example, the W7-X coils limit the size and shape of the NBI heating system duct, ultimately limiting the NBI pulse length [ Rust et al., 2011 ] . Arguably, the choice of MS will heavily depend on the design of the stellarator — conversely, the stellarator design will depend on the choice of the MS. It is therefore important to develop tools to optimize for good plasma accessibility for each possible MS, and study the trade-off between the MS, other plasma objectives and engineering constraints, such that the right MS can be chosen for the right application. Earlier work on stellarator accessibility focused on the engineering challenges of designing MS for a specific reactor design. For the HELIAS-5B design, a MS involving an in-vessel crane that would bring plasma-facing components to various ports was considered [ Schauer et al., 2013 ] . In the case of the ARIES-CS stellarator, a port-based MS was considered initially [ Brown et al., 2015 ] . In a second study, coils were optimized under the constraint to be straight on the outboard side of the device using the COILOPT++ code [ Gates et al., 2017 ] , allowing segments of the vessel to be moved radially to gain access to the device. These studies provided the best possible MS given a stellarator design; only recently were considered the reverse problem, were stellarators were optimized provided a MS. For example, Lion [2023] used the PROCESS code to explore the trade-off between vertical access ports and other engineering constraints and physics objectives. To the authors knowledge, there are however no prior attempts to explicitly target access port size in a stellarator optimization. Among the large parameter space of possible stellarators [ Giuliani et al., 2022 , Giuliani, 2023 , Giuliani et al., 2023 ] , some devices have good accessibility, indicating that it may be possible to design accessible stellarator reactors. Whether or not a device is accessible can be hard to assess — here we loosely define accessible as devices with large space between coils. Good examples are the Columbia Non-Neutral Torus (CNT) [ Pedersen et al., 2004 ] , or more recently the Compact Stellarator with Simple Coils (CSSC) [ Yu et al., 2022 ] , or the Columbia Stellarator eXperiment (CSX) [ Baillod et al., 2025 ] currently being designed at Columbia University. None of these examples include all of the metrics required for a reactor though. Another example is the Eos neutron source stellarator design [ Gates et al., 2025 , Swanson et al., 2025 , Kruger et al., 2025 ] , where an array of non-encircling windowpane coils shapes the plasma, while toroidal field coils are planar

Read the complete text