Quasi-symmetric error field correction and applications to ITER
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Quasi-symmetric error field correction and applications to ITER".
Paper summary: Kai: So, looking at the full picture of "Quasi-symmetric error field correction and applications to ITER," what’s the bigger picture for us working on experimental devices?
Mira: It boils down to realizing that those small, unavoidable three dimensional magnetic fields in a tokamak aren't just nuisance errors; they are part of a larger torque problem involving both immediate resonance risks and slower, continuous nonresonant torques <ref:2609.29849#pg1>.
Lev: The authors introduce this quasi-symmetric error field correction scheme as a way to handle that whole picture by minimizing the total torque response from the start, rather than just chasing one type of perturbation <ref:2609.29849#pg3>.
Kai: The title itself, "Quasi-symmetric error field correction and applications to ITER," tells you exactly what they're doing: they are applying a method designed for magnetic perturbations to the actual errors we see in ITER <ref:2609.29849#pg1>.
Mira: They show that this QS EFC scheme consistently yields low-torque solutions while strongly suppressing resonant response, outperforming single-mode resonant overlap EFC in most cases and approaching the performance of multimodal resonant EFC <ref:2609.29849#pg1>.
Lev: And for implementation, they point toward a practical pathway involving low-current identification combined with this quasi-symmetric error field correction <ref:2609.29849#pg1>.
Kai: It sounds like the key is using the remaining degrees of freedom in our control system to reduce that continuous torque penalty once we've met the threshold for resonance <ref:2609.29849#pg1>.
Mira: Exactly. They show that even with this higher-order nature, the QS EFC solution varies only tolerably between half- and full plasma current scenarios <ref:2609.29849#pg1>.
Lev: So, the implication is that we can build a system where the optimization naturally handles both immediate stability concerns and long-term confinement issues simultaneously <ref:2609.29849#pg3>.
Conclusion: Kai: So, to wrap up this paper on quasi-symmetric error field correction and applications to ITER, they are basically showing how you can fix those small three dimensional magnetic errors that happen in a tokamak by looking at torque instead of just resonance.
Mira: Right, it’s about using this systematic scheme to tackle both the immediate stability risks from resonant fields and the slower confinement problems caused by nonresonant torques like neoclassical toroidal viscosity.
Lev: From an engineering standpoint, they’re proposing a way to minimize that total torque response by using a torque-coupling matrix, which lets you figure out exactly what currents to apply based on the error field sources.
Kai: And the results show this quasi symmetric approach actually outperforms single mode resonant overlap correction in most situations, and it gets close to multimodal resonant correction performance.
Mira: That means they found a way to control the residual error fields better than just focusing on the strongest resonances alone, which is a big deal for ITER.
Lev: The authors also showed that this solution remains pretty robust; even if you change the plasma current from fifteen mega amperes to seven point five mega amperes, the correction currents don't need a major overhaul.
Kai: So what this means for us is that we can treat those continuous torque penalties more seriously alongside the immediate resonance constraints when designing control systems for devices like ITER.
Mira: It suggests a practical path forward where you identify low current conditions and then apply this quasi symmetric error field correction to manage the remaining nonresonant stuff.
Lev: They point to the vanishing first order torque gradient at this optimal point as a way to make that reduced sensitivity local, which is a really useful concept for real hardware design.
Kai: It seems like they’re giving us a clearer map for how these different types of magnetic imperfections actually couple into the overall plasma dynamics.
Seoul National University · ITER Organization
physics.plasm-ph
Submitted: 2026-09-24
Updated: 2026-11-17
Comments: 17 pages, 18 figures. Submitted to Nuclear Fusion. v2: torque minimization reformulated with fixed source amplitudes (Eq. 4); figures and references updated
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 91/100
The gist: Reliable correction of nonaxisymmetric error fields (EFs) is essential to the safety and performance of tokamak operation.
Key concepts
- Error Field Correction (EFC)
- EFC uses extra coils and fields to compensate for small, unwanted three-dimensional magnetic fields (error fields) that break the tokamak's symmetry. This is crucial because even tiny errors can trigger dangerous instabilities or degrade performance.
- Torque-Coupling Matrix Formalism
- This mathematical tool maps which specific error field sources correspond to which correction coils and their polarities. By using this matrix, researchers can determine the optimal distribution of correction currents needed to minimize the total integrated neoclassical torque.
- Quasi-Symmetric EFC (QS EFC)
- This strategy minimizes the total neoclassical toroidal viscosity (NTV) torque across all error field sources at once, rather than just focusing on one specific resonance. It treats resonant and nonresonant effects equally, leading to a solution that is robust and performs well against various intrinsic error field scenarios.
Terminology
Summary
Reliable correction of nonaxisymmetric error fields (EFs) is essential to the safety and performance of tokamak operation. p. 0 p. 0 Here, a systematic EFC scheme is introduced that minimizes both resonant and nonresonant EF effects simultaneously based on neoclassical torque response computed from self-consistent perturbed equilibria. p. 0 This quasi-symmetric (QS) EFC scheme outperforms single-mode resonant overlap EFC in most cases and approaches the performance of multimodal resonant EFC. p. 0
Introduction to the Problem
Tokamak operation is highly sensitive to a small three-dimensional (3D) magnetic field that breaks toroidal symmetry. Even a perturbation at the level of δB/B = 10−3 ∼ 10−4 can trigger a dangerous resonant response such as locked modes or tearing modes. p. 1 These small 3D fields are inevitable in tokamaks due to imperfections in the construction and assembly of the primary magnets and surrounding machine structures. p. 1 Error-field correction (EFC) mitigates these risks by using additional non-axisymmetric coils and fields that can effectively compensate intrinsic EFs. p. 1
Torque Coupling Matrix Formalism
A torque-coupling matrix formalism is introduced to map EF provenance to actuator choice and polarity, enabling transparent actuator planning beyond resonance-only metrics. This matrix contains all the information needed for the purpose of NTV minimization in our framework. p. 2 The optimal distribution of I is uniquely determined as the eigenvector associated with the minimum eigenvalue. p. 2 The cross-term represents the interference between the intrinsic error field and the applied correction field. p. 2
Correction Strategies Comparison
Three correction strategies are compared: (i) overlap-optimal correction as the operational reference, (ii) a conditional resonance-optimal extension that minimizes the total residual resonant flux on selected n = 1 rational surfaces, and (iii) QS EFC based on the quasisymmetric magnetic perturbation (QSMP) concept which minimizes neoclassical toroidal viscosity (NTV) torque while retaining a penetration-relevant resonance safeguard. The QS EFC solution is determined by torque minimization rather than by an active resonance constraint. p. 8
Performance and Robustness Analysis
The comparison shows that QS EFC produces a smaller q = 2 residual than overlap correction for every PF source shown. The mean q = 2 residual of QS EFC lies three orders of magnitude above the resonance-optimal value. p. 9 Furthermore, correction currents optimized at reduced plasma current (Ip = 7. p. 85 MA) retain useful performance when transferred directly to the 15 MA baseline equilibrium without re-optimization. p. 4
Conclusion and Implications
The paper concludes that torque effects must be treated on an equal footing with resonance in large devices like ITER. QS EFC uses the remaining freedom to reduce the continuous torque penalty once the threshold-like resonant requirement is met. p. 9 The results support a practical ITER pathway of low-current identification combined with QS EFC. p. 8 The vanishing first-order torque gradient at a QS EFC optimum provides a local mechanism for this reduced sensitivity. p. 0 The data and optimization scripts supporting the findings are available from the corresponding author upon reasonable request.
--- Page 1 ---
Quasi-symmetric error field correction and applications to ITER G.-G. Seo1, J.-K. Park1∗, S. p. 0 J. Han1 and M. p. 0 Dubrov2 1Department of Nuclear Engineering, Seoul National University, Seoul 08826, Republic of Korea 2ITER Organization, France E-mail: jkpark@snu. p. 0ac.kr Abstract. p. 0 Reliable correction of nonaxisymmetric error fields (EFs) is essential to the safety and performance of tokamak operation. Although resonant error field correction (EFC) is well established, supported by an improved understanding of 3D plasma response, the residual fields left after resonant EFC remain an open question: they are expected to be predominantly nonresonant, yet can still degrade both confinement and stability. p. 0 Here we introduce a systematic EFC scheme that minimizes resonant and nonresonant EF effects simultaneously, based on neoclassical torque response computed from self-consistent perturbed equilibria. p. 0 The scheme extends the method developed for designing quasisymmetric magnetic perturbations [1] to the case where actual error fields are present. p. 0 Application to standard ITER target plasmas with a range of intrinsic EF scenarios demonstrates the advantages of this quasi-symmetric (QS) EFC scheme for controlling residual EFs. p. 0 QS EFC consistently yields low-torque solutions while strongly suppressing resonant response, outperforming single-mode resonant overlap EFC in most cases and approaching the performance of multimodal resonant EFC. p. 0 We also show that the QS EFC solution varies only tolerably between half- and full-Ip ITER scenarios, despite the greater sensitivity expected from its higher-order nature. p. 0 Keywords: error field correction, quasi-symmetric magnetic perturbation, neoclassical toroidal viscosity, ITER, plasma response Submitted to: Nuclear Fusion arXiv:2609. p. 029849v1 [physics. p. 0plasm-ph] 24 Sep 2026
--- Page 2 ---
QS EFC for ITER 2 1. Introduction Tokamak operation is highly sensitive to a small threedimensional (3D) magnetic field that breaks toroidal symmetry. p. 1 Even a perturbation at the level of δB/B = 10−3 ∼ 10−4 can trigger a dangerous resonant response such as locked modes or tearing modes [2–7]. p. 1 It can also fundamentally alter particle motion and reduce toroidal rotation [8–10], thereby degrading overall confinement and stability. p. 1 These small 3D fields are inevitable in tokamaks, due to imperfections in the construction and assembly of the primary magnets and surrounding machine structures. p. 1 In particular, misalignments in TF/PF/CS coils such as millimeterlevel shifts and tilts, or coil windings and leads, can intrinsically produce such a level of 3D error fields, risking tokamak performance and operation. p. 1 Error-field correction (EFC) mitigates these risks by using additional non-axisymmetric coils and fields that can effectively compensate intrinsic EFs [2609. p. 1298495,11,12]. Because EFC is actuator-limited, the choice of how EF “severity” is quantified becomes decisive, because the adopted metric effectively dictates which spectral structures the limited actuators will prioritize. p. 1 This quantification has been progressively improved and applied to the International Thermonuclear Experimental Reactor (ITER) EFC strategy. p. 1 Earlier practice employed the “3-mode” criterion [14, 15] based on vacuum superposition in the estimation of resonant fields, providing useful guidance but not always isolating the plasma response structures most relevant to resonant drive and locking probabilities. p. 1 Recent approaches instead emphasize resonant physics via plasma response modeling and coupling matrices that connect external 3D fields to shielded resonant flux on selected rational surfaces [2609. p. 1298495,11]. In this framework, Singular Value Decomposition (SVD) schemes systematically identify dominant response structures and define device-agnostic overlap measures for ranking locking-relevant content. p. 1 While holding this scheme as a robust reference, the focus of EFC has shifted from dominant-structure targeting (overlap), to multisurface resonance minimization [17], and ultimately to formulations that explicitly incorporate non-resonant physics such as NTV torque. p. 1 A central issue is that resonant minimization alone does not uniquely control the residual 3D field spectrum. p. 1 Different coil solutions can achieve comparable resonant suppression while leaving substantially different non-resonant fields, which can still drive neoclassical toroidal viscosity (NTV) braking and reduce toroidal rotation. p. 1
--- Page 3 ---
QS EFC for ITER 3 2. Torque-coupling matrix for extended EFC Conventional error field correction (EFC) has been developed primarily to suppress resonant magnetic perturbations, motivated by their immediate impact on disruptive response including locked modes [2609. p. 2298495,11,13]. Two issues arise, however, when EFC is extended to include nonresonant perturbations. p. 2 One approach is to minimize the total NTV torque integrated over the plasma. p. 2 This integrated torque τφ is expressed as τφ(I) = I†TI, (1) where T represents the contribution of each correction and source, as well as their nonlinear coupling, to the integrated torque. p. 2 Once the torque-coupling matrix is fixed, the optimal distribution of I is uniquely determined as the eigenvector associated with the minimum eigenvalue. p. 2 The quantity (Tcs)c,s thus plays the role of a complex torque coupling from error-field source s to correcting coil c. p. 2 p. 2
--- Page 4 ---
QS EFC for ITER 4 Figure 1 Full torque-coupling matrix T for the 15 MA ITER baseline scenario, evaluated for unit shifts of the TF/PF/CS coils as EF sources and for the six correcting-coil families EFCB/EFCM/EFCT/RMPL/RMPM/RMPU. The augmented vector I = [Ic; Is] contains both the correcting-coil currents and the fixed error-field amplitude. p. 3 The quantity (Tcs)c,s thus plays the role of a complex torque coupling from error-field source s to correcting coil c.
--- Page 5 ---
QS EFC for ITER 5 Figure 2 Torque-coupling matrix evaluated in the EFC-only actuator space for the 15 MA baseline equilibrium. Panel (a) uses a signed normalization to the largest absolute real element of the selected actuator matrix; panel (b) shows the EFCB, EFCM, and EFCT actuator geometry. p. 4
--- Page 6 ---
QS EFC for ITER 6 Figure 1 Full torque-coupling matrix T for the 15 MA ITER baseline scenario, evaluated for unit shifts of the TF/PF/CS coils as EF sources and for the six correcting-coil families EFCB/EFCM/EFCT/RMPL/RMPM/RMPU. Figure 4 is the case most relevant to error-field correction because it explicitly includes an intrinsic error-field source in the augmented representation. p. 3 The minimumeigenvalue direction of the full matrix identifies the joint torque-favorable direction, and normalizing this direction to (Is)PF1 = 1 directly yields the correcting-coil current amplitudes and phases required to minimize torque in the presence of that error field. p. 3
--- Page 7 ---
QS EFC for ITER 7 Figure 7 Magnitude of the intrinsic n = 1, q = 2 resonant response in the 15 MA baseline equilibrium, normalized per centimetre of rigid coil shift. In the primary comparisons, we restrict the actuator set to the three EFC coil families EFCB, EFCM, and EFCT. p. 6 The first strategy is overlap-optimal correction [11, 13], which targets the dominant resonant structure identified by the SVD of the resonant coupling matrix. p. 6 The second strategy is resonance-optimal correction, which conditionally extends overlap-optimal correction. p. 6 The third strategy is QS EFC based on the quasisymmetric magnetic perturbation (QSMP) concept introduced in Sec. p. 7 p. 7 2.
--- Page 8 ---
QS EFC for ITER 8 Figure 10 Schematic illustration of resonance-optimal correction for the same PF4-tilt source shown in Fig. 9. The third strategy is QS EFC based on the quasisymmetric magnetic perturbation (QSMP) concept introduced in Sec. 2. In every case reported here, this constraint remained inactive because the unconstrained torque-minimizing solution already produced a residual resonance below this level. p. 8
--- Page 10 ---
QS EFC for ITER 10 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source. p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8
--- Page 11 ---
QS EFC for ITER 11 Figure 12 NTV-torque correction performance in the 15 MA baseline equilibrium for the same 1 cm PF shifts and strategies as Fig. 7. QS EFC gives the lowest residual for every tested PF source.
--- Page 15 ---
QS EFC for ITER 15 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 16 ---
QS EFC for ITER 16 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 17 ---
QS EFC for ITER 17 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 18 ---
QS EFC for ITER 18 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 19 ---
QS EFC for ITER 18 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 20 ---
QS EFC for ITER 18 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 21 ---
QS EFC for ITER 18 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
--- Page 24 ---
QS EFC for ITER 18 Figure 13 Radial accumulation of the GPEC integrated NTV torque, T(< ψN), for a 1 cm PF1 shift in the 15 MA baseline equilibrium. QS EFC gives the lowest residual for every tested PF source.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved system can perform systematic error field correction by minimizing both resonant and nonresonant effects simultaneously, based on neoclassical torque response computed from self-consistent perturbed equilibria.
-
The improved system can utilize a torque-coupling matrix formalism to map error field provenance to actuator choice and polarity, enabling transparent actuator planning beyond resonance-only metrics.
-
The system can compare three correction strategies: overlap-optimal correction, conditional multimodal resonance-optimal extension, and quasi-symmetric EFC based on the QSMP concept.
-
The improved system can evaluate transfer robustness by comparing correction currents optimized at reduced plasma current to the 15 MA baseline without re-optimization, assessing whether a practical workflow is feasible.
Related papers
- Implementation and verification of a new free boundary module for Resistive Wall Modes simulations with the nonlinear 3D full MHD code SpeCyl
- System Size Dependence of Collisionless Reconnection Rate
- SPECTRE: A robust solver for 3D equilibria with arbitrary topology
- High-quality electron beam generation from laser wakefield accelerators for driving compact free electron lasers
- Deflation Techniques for Stellarator Equilibrium and Optimization
- Rapid Vacuum Response Calculations for Non-axisymmetric Plasma Geometries