Design-Space Dominance Map
Find the active optical-link constraint before running high-fidelity simulations.
Enter SI engineering inputs. Defaults are pre-filled from the selected scenario. Enable "sweep this input" to mark a field for the 2D dominance-map sweep.
Ten dimensionless burden ratios rho_* produced by the projection x -> z. rho ~ 1 marks the engineering boundary.
Layer contributions
Each reduced invariant rho_* is attached to one active-branch layer; the engineering names below are the public mapping.
Burden vector z (radar)
log-scaled radius, clipped to z in [0, 10].
Pick two sweep axes from the design vector; the other inputs stay at their Design-Space form values. Renders the dominance map (dominant limiter) and the continuous closure-margin G heatmap side by side.
Seeded Monte Carlo over per-input uncertainties; reports P(closure margin G >= 0), sigma_G_dB, percentile band, and the dominant-limiter share. Same seed always produces the same result.
Per-input fractional sigma
Grid points within 1 dB of the closure boundary G = 0, or in a mixed-limiter window. These are the scenarios most worth following up with high-fidelity simulation.
Top-ranked scenarios deserving a high-fidelity follow-up, with a recommended model class for each. The priority score combines closure-margin proximity, mixed-boundary windows, and dominant-limiter category.
Local analytic test results plus a link to the lab-wide validation page.
Reproducible certificate. Same inputs + same version + same constants + same formulas always produce the same result_hash. The receipt_hash adds the export timestamp.
Show active-branch mapping (research id -> public engineering name)
Public engineering surrogate - constraint-compression framework - not mission-certified.