Zenodo — software record

SCPN Control: Neuro-Symbolic Stochastic Petri Net Controller for Plasma Control

· Miroslav Šotek, Michal Reiprich

tokamak controlstochastic Petri netspiking neural networkplasma physicsKuramoto oscillatorsadaptive couplingH-infinity controlLyapunov stability

Abstract

Standalone neuro-symbolic control engine that compiles Stochastic Petri Nets into spiking neural network controllers with formal verification guarantees. Features real-time adaptive Kuramoto coupling driven by tokamak diagnostics (beta, disruption risk, Mirnov, coherence PI), Lyapunov stability guard, H-infinity observer, plasma-native 8-layer Knm hierarchy, and WebSocket phase streaming. 52 Python modules, 5 Rust crates, 1888 tests.

Links

View DOI → Zenodo deposit CITATION.cff

Evidence bundles

The following bundles are linked from this publication. Each one carries a studio.*.v1 envelope with claim boundary, evidence kind, and empirical level.