direction_interpolation: cyclic linear interpolation across 10-degree sectors
spatial_interpolation: bilinear x/y interpolation of atlas to WTG positions
absolute_resource_source: selected-year calibrated chronological hub-height wind series
roughness_map: not yet used; inherited uniform grass atlas assumption
COP30 terrain correction applied.
The 36-sector WindNinja transfer atlas has been interpolated to every WTG and every hour of the selected chronological wind series.
Source Layout / Siting #46
Terrain atlas source #54
Hours / WTG 8784 / 15
Directions 36 × 10.0°
Reference gross AEP 238.880 GWh
Terrain-adjusted gross AEP 239.539 GWh
Terrain gross ΔAEP +0.659 GWh
Terrain gross ΔAEP +0.276%
WTG S range 0.9960 … 1.0060
WTG max |ΔWD| 0.261°
v0.14.0 integration gate.
Terrain flow is now applied to the chronological background wind, but this result is not yet offered to PyWake.
The next gate will verify PyWake consumption of the per-WTG hourly field while keeping StraightDistance unchanged.
Every completed calculation exports an XLSX workbook whose first worksheet is Cover.
The Cover is written in English and documents purpose, inputs, method, outputs, limitations and source/documentation links.
Calculation logic / How this works
Separate the absolute calibrated wind resource from local terrain transfer. Bypass remains the production baseline. v0.10.2 adds an executable WindNinja COM zero-relief validation gate before any real-terrain transfer is allowed.
Select a completed Turbine Definition & Gross AEP job and inherit its project/resource lineage and turbine hub height.
Default Bypass mode keeps the calibrated wind field uniform and therefore keeps terrain-adjusted gross AEP equal to the upstream reference gross AEP.
WindNinja COM — Flat DEM validation creates a constant-elevation synthetic DEM around the project reference point and runs 36 domain-average sectors from 0° to 350°.
Use 10 m/s at the same inherited hub height for WindNinja input and output, then verify that the interior domain preserves speed (S≈1) and direction (ΔWD≈0°).
After the flat PASS, run the synthetic north-south Gaussian ridge gate. Cross-ridge inflow must accelerate over the crest, while along-ridge flow stays close to uniform and the far field recovers near S=1.
Both synthetic validation modes are diagnostic only: they do not modify AEP and are intentionally not offered as PyWake wind-flow sources.
Real COP30 mode now derives a 36-sector relative speed-up/direction-turning atlas over the Layout-defined target area, using an additional solver boundary buffer and sector normalization to the calibrated project reference point.
COP30 precision check runs four directions at both 10 and 30 m/s on one DEM, reports numerical consistency and separate resolution evidence, and leaves the production input speed unchanged.
GWA4 remains an optional future relative spatial cross-check, not the absolute project wind climate.
Bypass, WindNinja flat-DEM validation, synthetic-ridge validation and the experimental real COP30 transfer-atlas stage are executable. The COP30 atlas is not yet applied to hourly AEP/PyWake; NinjaFOAM/OpenFOAM RANS and GWA4 remain later validation/cross-check stages. No paid terrain-flow dependency is part of the roadmap.