Pith. sign in

REVIEW 2 major objections 5 minor 51 references

Towards joint optimization of stellarator coils and support structures

T0 review · 2 major / 5 minor · reviewed 2026-07-11 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Jointly optimizing stellarator coil shapes and clamp locations with differentiable FEA cuts RMS von Mises stress by 2.4× at similar field error.

desk verdict Solid methods paper: first AD joint coil–clamp optimization with clean controls and open code; the 2.4× stress drop is real inside a simplified spring-clamp model, not a reactor prediction. read the letter →

arxiv 2607.05749 v1 pith:IJCLD4M2 submitted 2026-07-07 physics.plasm-ph

Lanke Fu , Alan A. Kaptanoglu This is my paper · ORCID
classification physics.plasm-ph
keywords stellaratorcoiloptimizationsupportstructuresfiniteelementanalysisdifferentiablemechanicsvonMisesstresshigh-temperaturesuperconductorlinearelasticity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Stellarator coils must make a precise magnetic field while surviving huge Lorentz and thermal loads, and their supports are usually designed by hand after the coils are already fixed. This paper argues that the support clamps themselves should be free variables inside the same optimization loop as the coil geometry. The authors supply coil-fem, a differentiable finite-element model that turns clamp positions and coil Fourier coefficients into stresses and deformations that can be differentiated and therefore optimized together. On a simplified W7-X-like set the joint run produces coils whose RMS von Mises stress is 2.4 times lower than the baseline with fixed top-and-bottom clamps, while the magnetic field error stays comparable. The result shows that where you clamp a coil can matter more for peak stress than modest changes to the coil shape alone, and that structural FEA can sit inside gradient-based stellarator design rather than only as a post-processing check.

What carries the argument

coil-fem: a fully differentiable finite-element pipeline that builds a finite-build coil mesh from a centerline, applies a spring-foundation boundary condition whose support patches move with optimizable clamp angles, solves linear elasticity, and returns gradients of stress and displacement with respect to both geometry and clamp locations.

What would settle it

Rebuild the optimized coil-and-clamp geometry in a conventional FEA code that includes a realistic cage or shell model; if the reported 2.4 imes RMS stress reduction disappears or the optimal clamp locations move substantially, the claim does not transfer.

Watch

Extended reading notes

Core claim

When coil Fourier coefficients and the parametric locations of support clamps are optimized together under a differentiable FEA load penalty, the resulting coil set achieves roughly 2.4 times lower RMS von Mises stress than an unoptimized baseline that uses fixed top-and-bottom clamps, while magnetic field error remains comparable.

Load-bearing premise

The supports are only soft spring patches glued to the coil surface with a large fixed stiffness, not a real deformable cage or shell whose flexing and load paths could change where stress actually peaks.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper introduces coil-fem, an open-source JAX-FEM-based tool that embeds differentiable linear-elasticity FEA (with spring-foundation support BCs) into stellarator coil optimization. From filamentary coils it builds finite-build rectangular meshes, applies Lorentz (Landreman–Hurwitz self-force plus mutual), gravity and uniform thermal contraction, and differentiates von Mises stress and displacement with respect to both coil Fourier coefficients and clamp parametric locations. On a simplified W7-X-like set with two clamps per coil, four controlled optimizations (clamp-only, coil-only, joint, and Lorentz-force proxy) show that joint optimization (case C) yields ~2.4 imes lower RMS von Mises stress and comparable field error relative to a fixed top/bottom-clamp baseline, while force-proxy optimization alone does not reduce stress.

Significance. If the result holds under the stated model, this is the first gradient-driven joint coil–support optimization in the stellarator literature and supplies a concrete, reproducible demonstration that clamp placement is a first-order lever on coil stress—something analytic force/torque proxies cannot capture. The open-source AD pipeline, DOLFINx cross-checks (pointwise agreement ~1e-8 when Bself is identical; RMS metrics within ~2%), and the A–D ablation suite are genuine strengths that lower the barrier to including structural FEA inside coil design loops. The work is therefore a useful proof-of-concept even though the support model is deliberately simplified.

major comments (2)
  1. §2.3, Eqs. (2) and (10): the central quantitative claim (2.4× RMS stress reduction) is obtained under isotropic spring-foundation patches with a single large fixed k0. The paper itself states that this BC “cannot accurately predict flexing in the support cage.” Because real cage compliance would redistribute loads and could move the optimal clamp locations, the reported factor should be presented strictly as a result inside the two-clamp homogeneous-coil model, and a short sensitivity study (or explicit caveat in the abstract/conclusions) is needed before the number is treated as transferable.
  2. §3.2, Eq. (11) and Table 3: the multi-objective weights, k0 = 10^10 N m^{-3}, Ncl = 2 and n = 80 are free hyperparameters. While the A–D suite cleanly isolates the effect of optimizable clamps, the manuscript does not show that the 2.4× reduction is robust to modest changes in these choices. A brief robustness check (or an explicit statement that the factor is weight- and k0-dependent) would strengthen the load-bearing claim.
minor comments (5)
  1. Abstract and §3.2: “unoptimized baseline” is slightly ambiguous; clarify that the baseline uses the original W7-X centerlines with fixed top/bottom clamps (distinct from the real W7-X support inventory).
  2. Fig. 1 caption and §3.1: the n = 275 data point is referenced but the main optimization uses n = 80; a short note on why the lower resolution is adequate for gradients would help.
  3. Table 1 and §2.1: the fixed-mesh-topology assumption is listed; a sentence on how large geometry changes are prevented (or remeshed) during multi-grid Fourier optimization would improve reproducibility.
  4. Eq. (10): the logistic sigmoid transition width ϵ_cl is never given a numerical value; please state the value used.
  5. References: Kaptanoglu 2026 and related arXiv preprints are cited; ensure final DOIs/versions are updated if available at publication.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: joint coil-support optimization reports an empirical multi-objective result, not a quantity forced by definition or self-citation.

full rationale

The paper's central claim (case C: joint optimization of coil Fourier coefficients and clamp locations yields ~2.4× lower RMS von Mises stress at comparable field error versus a fixed top/bottom-clamp W7-X-like baseline) is an empirical outcome of a controlled A–D optimization suite, not a tautology. Field error JB (normalized squared flux on the plasma boundary) and Jload (volume integral of σv² from linear-elastic FEA) are independent external objectives; the composite penalty (Eq. 11) does not force one from the other by construction. Clamp locations enter only through the spring-foundation BC (Eqs. 2, 10), which is a modeling choice, not a fit recycled as a prediction. Related coauthor work (Kaptanoglu 2026) is cited and explicitly distinguished (fixed supports vs optimizable clamps). DOLFINx benchmarks and open code further make the FEA self-contained. No equation reduces the reported stress reduction to a normalization identity, fitted parameter, or self-citation chain. Score 0 is therefore appropriate.

Assumptions & free parameters 5 free parameters · 6 assumptions · 2 invented entities

The central numerical claim rests on standard continuum mechanics plus several modeling choices that define the load path (spring clamps, homogeneous body, Landreman–Hurwitz self-force, fixed mesh). Hyperparameters (k0, clamp size, penalty weights) are set by hand. No new physical particles or forces are postulated; the invented objects are software/method constructions.

free parameters (5)
  • spring coefficient k0
    Fixed at 1e10 N/m^3 to approximate rigid clamps; not derived from cage stiffness and directly sets how strongly clamps pin the surface.
  • clamp half-width rcl and transition width epsilon_cl
    rcl=0.3 m chosen for the study; controls the spatial support of k(x) and thus stress concentrations.
  • multi-objective penalty weights in J (Eq. 11)
    Coefficients 1e12, 200, 100, 100, 10, 1000 and Jload scale 1e-18 are empirically chosen from typical term magnitudes; they shape the trade-off that produces the reported design.
  • number of clamps Ncl=2 per coil
    Chosen for simplicity, not matching real W7-X support topology; strongly affects achievable stress reduction.
  • optimization mesh resolution n=80
    Fixed resolution for all optimization FEA; coarser than the highest benchmark n and may under-resolve local peaks.
assumptions (6)
  • domain assumption Linear isotropic elasticity with small-strain additive thermal decomposition (Eqs. 1, 7–8)
    Standard for many coil FEA studies but excludes geometric nonlinearity and large thermal gradients; invoked throughout §2.2 and Table 1.
  • ad hoc to paper Spring-foundation BC is an adequate AD-compatible surrogate for cage clamps (Eqs. 2, 10)
    Authors note it cannot model cage flexing; load-bearing for all stress and clamp-location results (§2.3).
  • domain assumption Body force uses Landreman–Hurwitz self-force and filament mutual field; Fbody independent of displacement
    Standard efficiency approximation in coil literature; stated in §2.2 and Table 1.
  • domain assumption Homogeneous isotropic 316LN properties and uniform current density for the entire coil body
    Simplifies away winding pack vs case; §2.1 and Table 2; limits fidelity to real W7-X coils.
  • ad hoc to paper Fixed mesh topology during optimization; rotation-minimizing rectangular sweep for finite-build coils
    Enables AD and matrix reuse (§2.1); may constrain geometry changes and mesh quality.
  • standard math Standard stellarator filament penalties (JB, length, spacing, linking, curvature) as in prior coil optimization literature
    Appendix A; used as non-load terms in the joint objective.
invented entities (2)
  • coil-fem differentiable FEA pipeline independent evidence
    purpose: Generate finite-build mesh, apply clamp BC, solve/differentiate stress and deformation w.r.t. coil geometry and clamp angles
    Software method, not a physical entity; independent evidence is the open-source release and DOLFINx benchmarks.
  • Sigmoid-sum spring field k(x,{phi_i}) for movable clamps
    purpose: Make clamp locations continuous optimizable variables without changing sparse matrix structure
    Modeling device specific to this paper’s AD requirement (Eq. 10); validated only inside the same FEA setup.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards joint optimization of stellarator coils and support structures." pith.science (2026). https://pith.science/paper/IJCLD4M2

@misc{pith2026260705749,
  author       = {Pith},
  title        = {Pith review of: Towards joint optimization of stellarator coils and support structures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJCLD4M2}},
  note         = {Machine review of arXiv:2607.05749}
}
read the original abstract

The support structure is an integral part of the design of nuclear fusion reactors, especially 3D stellarator devices. A practical reactor's coils and support structures must have three competing qualities: an accurate magnetic field for good confinement, sufficient rigidity to protect the brittle high-temperature superconductor (HTS) from damage, and a simple geometry for low-cost construction. In existing devices, the coil geometry is often optimized without knowledge of the support structures' design and the coils' true stress and deformation. The support structures are then placed by hand through repeated finite element analyses (FEA) until engineering requirements are met. This makes the structural design of stellarator coil systems lengthy and labor-intensive. Using new developments in differentiable structural mechanics, we present coil-fem, an open-source software tool that integrates support differentiable FEA into the stellarator coil optimization loop. It enables the integrated optimization of coil geometry and support clamp locations to simultaneously reduce magnetic field errors and stresses in the coil body. We also present the first combined coil-support optimization in the stellarator literature. Using a penalty term based on coil-fem, we produced a coil set with 2.4x lower RMS von Mises stress and similar field error compared to an unoptimized baseline.

Figures

Figures reproduced from arXiv: 2607.05749 by the authors.

Figure 1
Figure 1. A comparison between the von Mises stress [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Left: the shell-type support structure of NCSX [35]. Right: the cage-type support structure of CFQS [9]. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Spring coefficient function k(x, ϕi) that emulates two support clamps on the top and bottom of a W7-X coil. Coil width w1, w2 0.2 m Young’s Modulus E 2.05 × 1011 Pa Poisson’s ratio ν 0.3 Density ρ (for gravity) 8000 kg/m3 Integral thermal contraction (−∆l/l) 0.29% [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The minimum, maximum and median point-wise errors in von Mises stress (left) and displacement (right) [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: The RMS von Mises stress and displacements from [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The forward-solve time of coil-fem and DOLFINx. The DOLFINx runs use 4 Intel Xeon Platinum 8592+ 64C CPUs. The coil-fem runs use one Nvidia L40S GPU. The coil optimization uses the penalty: J ≡ 1012 max(JB − JB,baseline, 0)2 + Jload + 200Jl + 100Jcc + 100Jcs + 10Jlink …
Figure 7
Figure 7. Figure 7: A comparison of σv (color) and support clamp locations (gray glyphs) across all 5 configurations. Here, case B and D have the same clamp locations as the baseline. In case A and C, the support clamp locations are optimized. This leads to substantial reductions in peak …
Figure 8
Figure 8. Figure 8: Global metrics of all 5 configurations. Note that cases A and C have significantly reduced [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: The values of σv in case C (optimizing both the clamp locations and coils) with (left) and without thermal contraction (middle). The right figure shows their differences. The contributions from thermal contraction (−∆l/l) are secondary but still significant [PITH_FULL…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 51 canonical work pages

  1. [1]

    Overview of Disruptions with JET-ILW

    S.N. Gerasimov, P. Abreu, G. Artaserse, et al. “Overview of Disruptions with JET-ILW”. In:Nuclear Fusion 60.6 (June 2020), p. 066028.ISSN: 0029-5515, 1741-4326.DOI: 10.1088/1741-4326/ab87b0.URL: https: //iopscience.iop.org/article/10.1088/1741-4326/ab87b0(visited on 02/16/2021)

    work page doi:10.1088/1741-4326/ab87b0.url: 2020

  2. [2]

    Significance of MHD Effects in Stellarator Confinement

    A. Weller, S. Sakakibara, K. Y . Watanabe, et al. “Significance of MHD Effects in Stellarator Confinement”. In:Fusion Science and Technology50.2 (2006), pp. 158–170.DOI: 10.13182/FST06-A1231. eprint: https: //doi.org/10.13182/FST06-A1231.URL:https://doi.org/10.13182/FST06-A1231

    work page doi:10.13182/fst06-a1231 2006

  3. [3]

    Prospects for pilot plants based on the tokamak, spherical tokamak and stellarator

    J.E. Menard, L. Bromberg, T. Brown, et al. “Prospects for pilot plants based on the tokamak, spherical tokamak and stellarator”. In:Nuclear Fusion51.10 (Aug. 2011), p. 103014.DOI: 10.1088/0029-5515/51/10/103014. URL:https://dx.doi.org/10.1088/0029-5515/51/10/103014

    work page doi:10.1088/0029-5515/51/10/103014 2011

  4. [4]

    Engineering cost & schedule lessons learned on NCSX

    R L Strykowsky, T Brown, J Chrzanowski, et al. “Engineering cost & schedule lessons learned on NCSX”. In: 2009 23rd IEEE/NPSS Symposium on Fusion Engineering. San Diego, CA, USA: IEEE, June 2009

    work page 2009

  5. [5]

    ARIES-CS magnet conductor and structure evaluation

    X R Wang, A R Raffray, L Bromberg, et al. “ARIES-CS magnet conductor and structure evaluation”. en. In: Fusion Sci. Technol.54.3 (Oct. 2008), pp. 818–837

    work page 2008

  6. [6]

    ARIES-CS coil structure advanced fabrication approach

    Lester M Waganer, Kevin T Slattery, John C Waldrop III, et al. “ARIES-CS coil structure advanced fabrication approach”. en. In:Fusion Sci. Technol.54.3 (Oct. 2008), pp. 878–889. 11

    work page 2008

  7. [7]

    Design and test of the support elements of the W7-X magnet system

    C Damiani, Maurizio Gasparotto, Bert Giesen, et al. “Design and test of the support elements of the W7-X magnet system”. In:21st IEEE/NPS Symposium on Fusion Engineering SOFE 05. Knoxville, TN: IEEE, Sept. 2005

    work page 2005

  8. [8]

    Engineering design and manufacturing of the modular coil system for the quasi-axisymmetric stellarator CFQS-T

    Hiroyuki Tanoue, Sho Nakagawa, Kazuki Nagahara, et al. “Engineering design and manufacturing of the modular coil system for the quasi-axisymmetric stellarator CFQS-T”. en. In:Fusion Eng. Des.212.114853 (Mar. 2025), p. 114853

    work page 2025

Show all 51 references
  1. [9]

    Preliminary design and analysis of the CFQS supporting structure

    Guozhen Xiong, Yuhong Xu, Akihiro Shimizu, et al. “Preliminary design and analysis of the CFQS supporting structure”. en. In:Fusion Eng. Des.160.112021 (Nov. 2020), p. 112021

    2020

  2. [10]

    Design of the superconducting encircling coils for Helios, the planar coil stellarator

    Jamal R Olatunji, Daniel Nash, Danis Fort, et al. “Design of the superconducting encircling coils for Helios, the planar coil stellarator”. en. In:Fusion Eng. Des.230.115894 (Sept. 2026), p. 115894

    2026

  3. [11]

    ParaStell: parametric modeling and neutronics support for stellarator fusion power plants

    Connor Moreno, Aaron Bader, and Paul Wilson. “ParaStell: parametric modeling and neutronics support for stellarator fusion power plants”. In:Front. Nucl. Eng.3.1384788 (Apr. 2024)

    2024

  4. [12]

    A novel discontinuous-Galerkin deterministic neutronics model for fusion applications: workflow for stellarator reactor design studies

    Timo J Bogaarts and Felix Warmer. “A novel discontinuous-Galerkin deterministic neutronics model for fusion applications: workflow for stellarator reactor design studies”. In:Nucl. Fusion66.4 (Apr. 2026), p. 046020

    2026

  5. [13]

    Bayesian methods for magnetic and mechanical optimization of superconducting magnets for fusion

    Sam Packman, Nicolò Riva, and Pablo Rodriguez-Fernandez. “Bayesian methods for magnetic and mechanical optimization of superconducting magnets for fusion”. en. In:J. Fusion Energy44.1 (Mar. 2025)

    2025

  6. [14]

    A proof-of-concept for automated AI-driven stellarator coil optimization with in-the-loop finite-element calculations

    Alan A Kaptanoglu and Pedro F Gil. “A proof-of-concept for automated AI-driven stellarator coil optimization with in-the-loop finite-element calculations”. In: (Mar. 2026). arXiv:2603.15240 [physics.plasm-ph]

    2026

  7. [15]

    Efficient calculation of self magnetic field, self-force, and self-inductance for electromagnetic coils with rectangular cross-section

    Matt Landreman, Siena Hurwitz, and Thomas M Antonsen. “Efficient calculation of self magnetic field, self-force, and self-inductance for electromagnetic coils with rectangular cross-section”. In:Nucl. Fusion65.3 (Mar. 2025), p. 036008

    2025

  8. [16]

    Strain optimisation for ReBCO high-temperature superconducting stellarator coils in SIMSOPT

    Paul Huslage, Elizabeth J Paul, Mohammed Haque, et al. “Strain optimisation for ReBCO high-temperature superconducting stellarator coils in SIMSOPT”. en. In:J. Plasma Phys.91.2 (Apr. 2025)

    2025

  9. [17]

    Reactor-scale stellarators with force and torque minimized dipole coils

    Alan A Kaptanoglu, Alexander Wiedman, Jacob Halpern, et al. “Reactor-scale stellarators with force and torque minimized dipole coils”. In:Nucl. Fusion65.4 (Apr. 2025), p. 046029

    2025

  10. [19]

    A non-planar ReBCO test coil with 3D-printed aluminum support structure for the EPOS stellarator

    Paul Huslage, Tristan Schuler, Pedro F Gil, et al. “A non-planar ReBCO test coil with 3D-printed aluminum support structure for the EPOS stellarator”. In:IEEE Trans. Appl. Supercond.36.3 (May 2026), pp. 1–5

    2026

  11. [20]

    Experimental results from early nonplanar NI-HTS magnet prototypes for the Columbia stellarator eXperiment (CSX)

    D Schmeling, M Russo, B T Gebreamlak, et al. “Experimental results from early nonplanar NI-HTS magnet prototypes for the Columbia stellarator eXperiment (CSX)”. In:IEEE Trans. Plasma Sci. IEEE Nucl. Plasma Sci. Soc.(2026), pp. 1–6

    2026

  12. [21]

    https://github.com/johnviljoen/ spineax

    Lanke Fu.coil-fem: A differentiable FEA toolkit for stellarator coils. https://github.com/johnviljoen/ spineax. GitHub repository, MIT License. 2026

    2026

  13. [22]

    JAX-FEM: A differentiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science

    Tianju Xue, Shuheng Liao, Zhengtao Gan, et al. “JAX-FEM: A differentiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science”. In:Computer Physics Communications(2023), p. 108802

    2023

  14. [23]

    Introduction to Finite Element Methods

    Carlos A. Felippa. “Introduction to Finite Element Methods”. In: Lecture notes, Aerospace Engineering Sciences. University of Colorado, Boulder, CO, 2004. Chap. 13

    2004

  15. [24]

    Stress singularities in classical elasticity–I: Removal, interpretation, and analysis

    G B Sinclair. “Stress singularities in classical elasticity–I: Removal, interpretation, and analysis”. en. In:Appl. Mech. Rev.57.4 (July 2004), pp. 251–298

    2004

  16. [25]

    Society for Industrial and Applied Mathematics, Jan

    Pierre Grisvard.Elliptic Problems in Nonsmooth Domains. Society for Industrial and Applied Mathematics, Jan. 2011

    2011

  17. [26]

    Robert D Cook, David S Malkus, Michael E Plesha, et al.Concepts and applications of finite element analysis. en. 4th ed. Nashville, TN: John Wiley & Sons, Oct. 2001

    2001

  18. [27]

    O. C. Zienkiewicz, R. L. Taylor, and J. Z. Zhu.The Finite Element Method: Its Basis and Fundamentals. 7th. Oxford: Butterworth-Heinemann, 2013.ISBN: 978-1-85617-633-0

    2013

  19. [28]

    VIPER: an industrially scalable high-current high-temperature superconductor cable

    Zachary Seth Hartwig, Rui F. Vieira, Brandon N. Sorbom, et al. “VIPER: an industrially scalable high-current high-temperature superconductor cable”. en. In:Prof. Hartwig(Oct. 2020).ISSN: 1361-6668.URL: https: //dspace.mit.edu/handle/1721.1/133134(visited on 04/28/2026)

    2020

  20. [29]

    Computation of rotation minimizing frames

    Wenping Wang, Bert Jüttler, Dayue Zheng, et al. “Computation of rotation minimizing frames”. en. In:ACM Trans. Graph.27.1 (Mar. 2008), pp. 1–18

    2008

  21. [30]

    Design and construction of coil supporting structure and cryostat vessel for LHD

    H Tamura, A Nishimura, S Imagawa, et al. “Design and construction of coil supporting structure and cryostat vessel for LHD”. In:Advances in cryogenic engineering. Springer, 1998, pp. 253–260. 12

    1998

  22. [31]

    Modular coil design developments for the National Compact Stellarator Experiment (NCSX)

    D Williamson, A Brooks, T Brown, et al. “Modular coil design developments for the National Compact Stellarator Experiment (NCSX)”. en. In:Fusion Eng. Des.75-79 (Nov. 2005), pp. 71–74

    2005

  23. [32]

    The helically symmetric experiment, (HSX) goals, design and status

    F Simon B Anderson, Abdulgader F Almagri, David T Anderson, et al. “The helically symmetric experiment, (HSX) goals, design and status”. en. In:Fusion Technol.27.3T (Apr. 1995), pp. 273–277

    1995

  24. [33]

    Engineering challenges on the low-aspect-ratio quasi-axisymmetric stellarator CFQS-T and its upgrade status toward the 1 T operation

    Hiroyuki Tanoue. “Engineering challenges on the low-aspect-ratio quasi-axisymmetric stellarator CFQS-T and its upgrade status toward the 1 T operation”. In:Proceedings of the 25th International Stellarator-Heliotron Workshop (ISHW-2026). National Institute for Fusion Science (...

    2026

  25. [34]

    Stochastic anda posteriorioptimization to mitigate coil manufacturing errors in stellarator design

    Florian Wechsung, Andrew Giuliani, Matt Landreman, et al. “Stochastic anda posteriorioptimization to mitigate coil manufacturing errors in stellarator design”. In:Plasma Phys. Control. Fusion64.10 (Oct. 2022), p. 105021

    2022

  26. [35]

    Williamson.NCSX Modular Coils Failure Mode and Effect Analysis (FMEA)

    D. Williamson.NCSX Modular Coils Failure Mode and Effect Analysis (FMEA). Tech. rep. NCSX-FMEA- 140-02-00. Prepared by D. Williamson (WBS 14 Manager); Checked by M. Cole (WBS 1/Design Integration Manager); Approved by P. Heitzenroeder (Engineering Manager). Princeton, NJ: Prin...

    2008

  27. [36]

    Design and construction of WENDELSTEIN 7-X

    M Wanner, J-H Feist, H Renner, et al. “Design and construction of WENDELSTEIN 7-X”. en. In:Fusion Eng. Des.56-57 (Oct. 2001), pp. 155–162

    2001

  28. [37]

    Mechanical design and construction qualification program on ITER correction coils structures

    A Foussat, Wu Weiyue, Wei Jing, et al. “Mechanical design and construction qualification program on ITER correction coils structures”. en. In:Nucl. Eng. Des.269 (Apr. 2014), pp. 116–124

    2014

  29. [38]

    DOLFINx: the next generation FEniCS problem solving environment

    Igor A. Baratta, Joseph P. Dean, Jørgen S. Dokken, et al. “DOLFINx: the next generation FEniCS problem solving environment”. In:preprint(2023).DOI:10.5281/zenodo.10447666

    2023 doi

  30. [39]

    Adams, et al.PETSc Web page.http://www.mcs.anl.gov/petsc

    Satish Balay, Shrirang Abhyankar, Mark F. Adams, et al.PETSc Web page.http://www.mcs.anl.gov/petsc. 2015.URL:http://www.mcs.anl.gov/petsc

    2015

  31. [40]

    https: //developer.nvidia.com/cudss

    NVIDIA Corporation.NVIDIA cuDSS: A High-Performance CUDA Library for Direct Sparse Solvers. https: //developer.nvidia.com/cudss . Version 0.7.0. Documentation: https://docs.nvidia.com/cuda/ cudss/. 2025

    2025

  32. [41]

    https://github.com/johnviljoen/spineax

    John Viljoen.Spineax: Sparse Linear Solvers in JAX. https://github.com/johnviljoen/spineax. GitHub repository, MIT License. 2025

    2025

  33. [42]

    New method to design stellarator coils without the winding surface

    Caoxiang Zhu, Stuart R Hudson, Yuntao Song, et al. “New method to design stellarator coils without the winding surface”. In:Nucl. Fusion58.1 (Jan. 2018), p. 016008

    2018

  34. [43]

    Augmented Lagrangian methods produce cutting-edge magnetic coils for stellarator fusion reactors

    Pedro F Gil, Weiping Li, Julianne Stratton, et al. “Augmented Lagrangian methods produce cutting-edge magnetic coils for stellarator fusion reactors”. In: (July 2025). arXiv:2507.12681 [physics.plasm-ph]

    2025 arXiv

  35. [44]

    Automated optimization of stellarator coils

    Michael Drevlak. “Automated optimization of stellarator coils”. en. In:Fusion Technol.33.2 (Mar. 1998), pp. 106–117

    1998

  36. [45]

    W7- X VMEC equilibrium database, configuration 1000_1000_1000_1000_+0000_+0000/01/00jh_l; β= 0 , zero net toroidal current

    Joachim Geiger.Wendelstein 7-X Standard Configuration (EIM) Free-Boundary VMEC Equilibrium. W7- X VMEC equilibrium database, configuration 1000_1000_1000_1000_+0000_+0000/01/00jh_l; β= 0 , zero net toroidal current. Max Planck Institute for Plasma Physics (IPP), Greifswald. 20...

    2018

  37. [46]

    Stellarator optimization for good magnetic surfaces at the same time as quasisymmetry

    Matt Landreman, Bharat Medasani, and Caoxiang Zhu. “Stellarator optimization for good magnetic surfaces at the same time as quasisymmetry”. en. In:Phys. Plasmas28.9 (Sept. 2021), p. 092505

    2021

  38. [47]

    SIMSOPT: A flexible framework for stellarator optimization

    Matt Landreman, Bharat Medasani, Florian Wechsung, et al. “SIMSOPT: A flexible framework for stellarator optimization”. In:J. Open Source Softw.6.65 (Sept. 2021), p. 3525

    2021

  39. [48]

    SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python

    Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, et al. “SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python”. In:Nature Methods17 (2020), pp. 261–272.DOI:10.1038/s41592-019-0686-2

    2020 doi

  40. [49]

    Proof of concept of a fast surrogate model of the VMEC code via neural networks in Wendelstein 7-X scenarios

    Andrea Merlo, Daniel Böckenhoff, Jonathan Schilling, et al. “Proof of concept of a fast surrogate model of the VMEC code via neural networks in Wendelstein 7-X scenarios”. In:Nucl. Fusion61.9 (Sept. 2021), p. 096039

    2021

  41. [50]

    Stellarator design exploration using symbolic- regression neutronics surrogates

    Enrique Miralles-Dolz, Michael Churchill, Jacob Schwartz, et al. “Stellarator design exploration using symbolic- regression neutronics surrogates”. In:IEEE Trans. Plasma Sci. IEEE Nucl. Plasma Sci. Soc.54.6 (June 2026), pp. 2879–2886

    2026

  42. [51]

    Low-dimensional geometry learning for turbulence prediction in optimized stellarators

    Xishuo Wei, Handi Huang, Haotian Chen, et al. “Low-dimensional geometry learning for turbulence prediction in optimized stellarators”. In: (Mar. 2026). arXiv:2603.17366 [physics.plasm-ph]

    2026 arXiv

  43. [52]

    Fu Lanke.Coil-fem Validation Dataset. 2026. 13

    2026

Pith tools

Reviewed July 11, 2026 · model on record in the stance chip above.