Methanol-Water (NRTL) VLE validation vs. experiment
within toleranceMaximaLabs reproduces the methanol/water vapor-liquid equilibrium under the nrtl property package and measures the deviation from published experimental data — live, on every page load, never a pinned number.
Model vs. experiment
The solid curve is the NRTL model; the amber markers are the cited experimental points.
Re-execute this validation on our server right now — it reproduces the published AAD exactly (the model is deterministic; only the wall-clock varies).
Reproduce this number
We publish our inputs, our numbers, and the citation so the comparison is reproducible. To benchmark another simulator, build the identical methanol/water case in it under an equivalent property method and compare against the cited reference — the ground truth both are measured against. We don't publish competitor numbers we didn't measure.
uv run python scripts/reproduce_validation.py # recomputes every number live