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

The Twin State Vector and Interfaces

What KRONOS-CTRL actually holds and publishes: a defined state vector with lineage that L1, MPC, and the anomaly ensemble consume.

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.

A defined state, not a black box

KRONOS-CTRL maintains an explicit, versioned state vector rather than an opaque model. It comprises the plasma equilibrium and stability margins, the neutron flux and breeding balance, the thermal and strain fields, and the electrical/DEC and grid-sync state, plus per-quantity uncertainty and a lineage tag pointing back to the L2 features and model versions that produced it.

Interfaces

The state vector is the contract between L3 and its consumers. L1 receives actuation intents and envelope constraints derived from it; MPC reads the current state and its forward projection as f_twin; the anomaly ensemble reads it to compute forecast residuals; operators read a curated view. Because it carries lineage, any decision can be traced to the features and model versions behind it, which is required for validation and post-event analysis.

The state is published on a fixed cadence with both a now estimate and the shadow projection, so consumers always know current state and near-future state together. Uncertainty travels with every quantity, so a downstream controller can see not just the value but how much to trust it and tighten its margins when trust is low.

Strict metadata governance from L2 carries through here: the twin state inherits the same lineage discipline, so the whole chain from analog port to control decision is auditable. This is what lets Kronos treat the twin as evidence, not just as an aid.

Content reviewed August 2026 · design-and-simulation stage