Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
AI Architecture › L3 · Twin Modeling & AI
L3 · Twin Modeling & AI

Free-Boundary Equilibrium and Coil Currents

The breeder's plasma boundary is not fixed; Kronos's free-boundary PINN co-solves the plasma equilibrium and the external coil currents that hold its shape.

THE STACK · click to jumpL7Ecosystem & StrategyL6Experience & VisualizationL5Applications & CopilotsL4OrchestrationL3Twin Modeling & AIL2Data FabricL1Control PlaneL0Foundation▲tlmctl▼L3 · TWIN MODELING & AIThe KRONOS-CTRL digital twin and its predictive shadow.1KRONOS-CTRL Twinlive plant state2GNNscoupled subsystems3PINNsphysics-constrained4Anomaly Ensemblesdrift & fault detection5MPCreceding-horizon control6Predictive Shadowruns seconds aheadMACHINE TIEState estimate descends to L1 control; alerts rise to L4 / L5.KRONOS FUSION ENERGYAI-NATIVE S.M.A.R.T. GENERATORTWIN MODELING & AISHEET 05REV. 2026-08L3 · AI-NATIVE STACK
L3 · Twin Modeling & AI — its place in the stack (left, click any layer) and its internal components (right). Telemetry rises; control descends.

Fixed vs free boundary

A fixed-boundary solve assumes the plasma edge is known and finds the interior flux. The real control problem is the opposite: the poloidal-field coil currents are the actuators, and the boundary shape is a consequence. A free-boundary PINN takes coil currents (and plasma profiles) as input and returns both the interior equilibrium and the resulting last-closed flux surface, including the X-point and strike points on the divertor.

Coupling to actuation

This is what makes the PINN useful to shape-control MPC. The controller proposes coil-current trajectories; the free-boundary PINN predicts the shape they produce, including whether the negative-triangularity delta -0.30 target and the divertor strike-point placement are met, all inside the shadow horizon. The PINN's differentiability means MPC can also obtain gradients of shape with respect to coil current directly, which sharpens the optimization.

python
# free-boundary map used by shape-control MPC
#   inputs : I_coils (PF/CS currents), p'(psi), FF'(psi)
#   output : psi(R,Z), boundary(theta), X-point (R_x,Z_x), delta, kappa
shape = pinn_freeboundary(I_coils, p_prime, ff_prime)
# MPC reads shape.delta, shape.strike_points and their d/dI_coils

The vacuum-vessel geometry matters here. The breeder's CrMoNbV vessel and the thermal expansion tracked by the Thermomechanics module change the coil-to-plasma coupling slightly as the machine heats; the free-boundary PINN is conditioned on the current vessel geometry from the twin so its shape predictions stay accurate through a pulse.

Because coil currents are physical actuators with limits, the free-boundary map is always used inside the certified safe operating envelope: MPC may only request shapes reachable with in-limit currents, and the failsafe path for coil faults remains an L1 hardware function independent of the PINN.

Content reviewed August 2026 · design-and-simulation stage