This paper introduces a novel, first-principles geometric framework for predicting tokamak plasma disruptions, a critical challenge for magnetic confinement fusion. By mapping the plasma's safety factor profile (q(r)q(r)) to a complex stability domain, the author derives a scale-invariant, device-independent metric called the Curvature–Modularity Correspondence (CMC) invariant (Λ⋅aΛ⋅a). This metric rigorously encodes the density of stabilizing magnetic shear without relying on arbitrary empirical cutoffs. Key Highlights: Geometric Stability Limit (GSL): The study uses variational calculus and MHD force balance to mathematically prove a strict lower bound for the invariant, acting as a fundamental topological trigger for magnetic reconnection. Experimental Validation: Tested on a multi-machine database of 715 discharges (JET, DIII-D, KSTAR), the metric achieves an AUC of 0.847 and a median warning time of 32 ms. Performance: It significantly outperforms traditional empirical indicators (like q95q95 and βNβN) and rivals state-of-the-art machine learning models, while remaining fully physically interpretable and independent of the specific tokamak device. Extended MHD: The framework is successfully generalized to multi-fluid regimes by deriving an effective safety factor (qeffqeff) that incorporates bootstrap current and two-fluid effects. Ultimately, the paper offers a universal, regularization-free, and mathematically rigorous disruption monitor for next-generation fusion reactors like ITER.
Sami I. Almuaigel (Mon,) studied this question.