SCPN-Fusion-Core

Neuro-symbolic control framework for tokamak fusion reactors

Version 3.9.11

SCPN-Fusion-Core is an open-source Python and Rust framework for control-first tokamak simulation. It helps teams express plasma-control logic as stochastic Petri nets, compile that logic into spiking-neural controllers, run it against physics-informed plant models, and publish reproducible validation evidence without hiding blocked full-fidelity gates.

What makes it different: Most fusion codes are physics-first (solve equations, then bolt on control). SCPN-Fusion-Core is control-first – it provides a contract-checked neuro-symbolic compilation pipeline where plasma control policies are expressed as Petri nets, compiled to stochastic LIF neurons, and executed against physics-informed plant models.

Note

Evidence boundary: This is not a replacement for TRANSP, JINTRAC, GENE, CGYRO, GS2, DREAM, Aurora, STRAHL, or EFIT. The native solver stack now exposes nonlinear 5D gyrokinetic state contracts, local electromagnetic diagnostics, decomposition contracts, and fail-closed full-fidelity benchmark gates, but production parity still requires same-case external reference outputs and quantitative thresholds. Treat the current public evidence as a control-algorithm, native-kernel, and validation-framework release, not as a completed production turbulence or reconstruction code.

Reader Orientation

SCPN-Fusion-Core is best read as three connected systems: a control compiler, a native physics-kernel laboratory, and a reproducible validation/reporting surface. The public reports distinguish validated local contracts from blocked full-fidelity parity rows. Use the project overview, onboarding guide, and benchmark taxonomy before treating any number as a production claim.

What Readers Can Do Today

  • Prototype controller and replay logic against local physics-informed models.

  • Inspect native Python and Rust solver kernels and benchmark reports.

  • Learn fusion-control architecture through quickstarts and notebook exports.

  • Prepare same-case reference-solver campaigns with explicit blockers.

  • Evaluate the funding and hardware needed for stronger parity evidence.

Licensing Model

SCPN-Fusion-Core is dual-licensed. The open-source path is AGPL-3.0-or-later for research, review, education, reproducible validation, public derivative work, and network-deployed AGPL services. Commercial licences are available for proprietary reactor programs, internal deployments, closed operational workflows, commercial support, and integrations that cannot use AGPL reciprocal terms.

Commercial licensing contact: protoscience@anulum.li

Key Features

  • Neuro-symbolic compiler – Petri net to SNN compilation with formal verification (37 hardening tasks)

  • Safety interlocks – inhibitor-arc hard-stop channels with contract-proof checks for thermal/density/beta/current/vertical limits

  • Grad-Shafranov equilibrium – Picard + Red-Black SOR or multigrid V-cycle, validated on 8 SPARC GEQDSK files

  • 1.5D radial transport – coupled energy/particle transport with IPB98(y,2) confinement scaling

  • AI surrogates – FNO turbulence, neural equilibrium, neural transport MLP, ML disruption predictor

  • Digital twin – real-time twin with RL-trained MLP policy and chaos monkey fault injection

  • Rust acceleration – 11-crate Rust workspace providing 10–50x speedups with pure-Python fallback

  • Real data validation – SPARC GEQDSK, ITER 15 MA baseline, ITPA H-mode confinement database

  • Fail-closed full-fidelity campaign – explicit GENE/CGYRO/GS2, DREAM, Aurora/STRAHL, FreeGS, electromagnetic, and decomposition blockers tracked as reports rather than promoted to passes

  • Graceful degradation – every module works without Rust, without SC-NeuroCore, without GPU

Getting Started

Public Documentation Hubs

Learning Path

Advanced Tutorials

User Guide

API Reference

Example Notebooks

Reference

Project

Indices and Tables