Enregistré dans:
| Auteur principal: | |
|---|---|
| Format: | Recurso digital |
| Langue: | |
| Publié: |
Zenodo
2026
|
| Sujets: | |
| Accès en ligne: | https://doi.org/10.5281/zenodo.19412716 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
Table des matières:
- <div> <div> <div> <div> <div> <div> <div> <div> <div> <div> <div> <div> <p>This paper proposes a fundamentally new approach to one of fusion energy's most critical unsolved problems: predicting when a tokamak plasma will disrupt before it happens. Current neural network predictors work well on the machines they are trained on but fail when transferred to new machines — and cannot be trained on ITER at all before ITER exists.</p> <p>The solution comes from mathematics. <a href="https://doi.org/10.5281/zenodo.19014729">Cross-Stitch Theory</a>, established in Cosentino (2026), shows that the Mathieu group M₁₂ — a rare sporadic symmetry group of order 95,040 — governs coherent magnetic energy release in plasma systems. Its connection to the binary Golay error-correcting code produces a universal resonance window: energy release occurs when a normalised field ratio falls within [0.85, 1.15]. In solar physics this has been validated by 787 predictions across a three day period. In this paper, the same window is mapped onto the tokamak safety factor condition q₉₅/q_rational ≈ 1.0.</p> <p>Testing this against 14 disruption-onset data points extracted from the landmark Hender et al. (2007) review, 93% fall within the window with a mean ratio of 0.999 ± 0.056 — pinned to within 0.1% of the predicted centre. An independent solar simulation study by Gyeltshen et al. (2026) then provides empirical calibration of four new M₁₂ constants, confirming every major CST operational rule from an entirely separate dataset.</p> <p>The paper derives a six-component machine-independent disruption state vector, states six falsifiable predictions testable on existing databases without new experiments, and specifies a four-stage validation protocol. Because the framework is derived from pure symmetry rather than machine-specific data, a predictor built on it can be deployed on ITER from first plasma — the defining advantage over every existing approach.</p> </div> </div> </div> </div> </div> </div> </div> </div> </div> <div> <div> <div> <div> <div> </div> </div> <div> </div> <div> </div> <div> <div> </div> </div> </div> </div> </div> </div> </div> <div> <div> <div> <div> </div> </div> </div> </div> </div>