[cmccomb]

Scope

Adds a full-duration physical comparison for issue #622. The test independently advances public Python OWC motion from the pinned HDF5 and PM wave phases, then compares body and piston motion, PTO force and power, air pressure and flow, and integrated work with a corrected MATLAB case.

The corrected case is derived from the original pinned Applications OWC/OrificeModel: a repository-local MATLAB script adds the existing piston reaction to the flexible GBM state-equation input on a disposable model copy. A refined solver clock of 0.00015625 s retains 0.005 s output and force-sampling clocks. The source fixture contains 26,001 output samples; work and maximum Mach use all 832,001 internal orifice samples. The follow-up Python-repository workflow in #641 regenerates this source from the pinned original case and gates fresh MATLAB output against independent Python.

Measured results

• Peak motion gaps: 2.246 mm surge, 0.116 mm heave, 0.359 mrad pitch, 0.022 mm flexible displacement.
• Peak speed gaps: 0.357 mm/s surge, 0.245 mm/s heave, 0.165 mrad/s pitch, 0.163 mm/s piston.
• Peak air/PTO gaps: 0.450 N force, 0.00229 kPa pressure, 0.0000320 m³/s flow, 0.0000405 kW power.
• Integrated work: Python 865.042 J versus MATLAB 864.422 J, a 0.0717% gap. Source peak Mach is 0.155, below the configured 0.3 warning threshold.

The test also reconstructs MATLAB orifice outputs from its separately logged piston speed and the documented air law. All declared limits pass. The repository-local 130 s source run reproduces the archived arrays within 3.96e-11 and passes the fresh-source Python gate.

Parity status

This is a derived-model comparison. The unchanged published Simulink input omits piston force from flexible state integration, so OWC/OrificeModel retains qualified_motion status and the strict badge count does not change.