A new computational approach could help reduce subtle but damaging magnetic imperfections in tokamak reactors such as ITER, according to a recent preprint (arXiv:2609.29849v1). The work addresses so-called error fields—small departures from the ideal, symmetrical magnetic geometry that confines the hot plasma—and proposes a way to correct both the well-known resonant components and the harder-to-handle nonresonant leftovers at once.

Error fields are a persistent practical problem for magnetic confinement fusion. Even small nonaxisymmetric deviations can interact with the plasma and produce torques or drive instabilities that degrade confinement and, in some cases, threaten safe operation. Existing correction schemes focus largely on eliminating resonant errors that line up with particular magnetic surfaces. But experiments and theory have increasingly shown that residual, nonresonant error fields can still matter for performance.

The new study develops a “quasi-symmetric” error-field-correction (QS EFC) scheme that is explicitly designed to minimize both resonant and nonresonant effects simultaneously. The authors build their method on physics-based calculations of the plasma’s neoclassical torque response, using self-consistent perturbed equilibria to predict how the plasma reacts to candidate correction fields. In other words, the correction is chosen with an eye to how the plasma itself will respond, not only to the imposed coil currents.

Applied in simulation to standard ITER target-plasma scenarios with a variety of assumed intrinsic error-field conditions, the QS EFC solutions consistently produce low net torque while strongly suppressing resonant responses. The paper reports that the quasi-symmetric approach outperforms single-mode resonant-overlap correction in most of the test cases, and comes close to the performance of more elaborate multimode resonant-correction strategies.