A neural RT-EFIT: reconstruct the equilibrium from magnetics inside the control loop.
The AI-Native S.M.A.R.T. Generator Master Blueprint — eight layers (L0→L7), one control stack, wired to both machines. Telemetry rises in microseconds; control descends the same path.
Category: B · physics · Plugs into: L1 / L3 · Horizon: FOAK · Status: on the roadmap — not yet built
What it is
The controller needs the plasma shape and equilibrium in real time; a neural real-time EFIT reconstructs them from magnetic measurements at kilohertz rates.
The method
A network trained to invert magnetics into the Grad–Shafranov equilibrium, faster than a classical solver, feeding the shape controller directly.
Why it matters
Real-time shape control — holding negative triangularity δ = −0.30 — needs a real-time equilibrium. This is the sensor-to-shape link the fast loop depends on. Plugs into L1/L3.