Phase 2 · fleet extensions

KHEAT

BUILT
heating&CD

heating & current-drive actuator-response surrogate — RandomForest over a 3888-point current-drive design scan; predicts driven current I_cd from RF/NBI drive parameters + plasma state (**R²=0.949**). Feeds KAIROS heating control. Reduced CD model — RF/NBI ray-tracing is the fidelity upgrade.

Headline benchmark — honest, computed
heating/current-drive surrogate: driven-current I_cd R²=0.949 over 3888 configs (reduced CD; RF/NBI ray-tracing = upgrade)

What it accelerates

KODEX KHEAT is a fast, calibrated surrogate for full RF/NBI ray-tracing (GENRAY/TORAY/NUBEAM) — reproducing its result at inference speed, so it runs inside a real-time control loop or a design search where the full computation is far too slow to call.

Provenance
SIM
De-risking gates
H10
Rollout
Phase 2 · Built

Use it

One line, one contract — a prediction, its uncertainty, and whether the input is in-domain.

from kronos_ml import KHEAT

model = KHEAT()                          # loads the trained surrogate
y, sigma, in_domain = model.predict(x)   # y = heating&cd prediction
if not in_domain:                       # out of its trusted region
    fall_back_to_full_physics()          # KHEAT abstains, never extrapolates

How to trust it

Every KODEX code wraps the shared spine — KHALO for calibrated uncertainty and KGATE for the out-of-domain gate — so it reports how confident it is and abstains rather than extrapolate. Benchmarks are computed on held-out data with a fixed seed; pre-registered misses are kept, not hidden.

In the fleet

Get it · cite it

Part of the open kronos-ml package (Apache-2.0). Open-access deposits with citable DOIs are listed below.

Open-access deposits, each with a citable DOI: