Sky Camera Calibration Certificate Generator (NIST-Traceable)
The Sky Camera Calibration Certificate Generator (NIST-Traceable) is an Excel-based computational resource that automates the generation of metrologically rigorous calibration certificates for all-sky (hemispherical) imaging systems used in solar forecasting. It integrates photogrammetric, radiometric, and geometric calibration workflows with NIST-traceable reference standards and uncertainty quantification. The tool outputs auditable, ISO/IEC 17025–aligned certificates compliant with renewable energy forecasting infrastructure requirements.
📖 Overview
📑 Key Components
🎯 Applications
- ✓ Validating sky cameras in utility-scale solar forecasting networks
- ✓ Supporting ISO/IEC 17025 accreditation for meteorological sensor labs
- ✓ Enabling traceable model input uncertainty quantification in AI solar nowcasting
📐 Key Formulas
Lens Distortion Correction (Brown-Conrady Model)
x_corrected = x_distorted * [1 + k1*r² + k2*r⁴] + 2*p1*x*y + p2*(r² + 2*x²); y_corrected = y_distorted * [1 + k1*r² + k2*r⁴] + p1*(r² + 2*y²) + 2*p2*x*y
Corrects radial and tangential distortion in fisheye sky images using polynomial coefficients derived from calibration target images.
Zenith Angle Mapping (Pinhole Projection with Fisheye Correction)
θ = 2 * arctan(r / (2*f)) * (1 + α₁*r² + α₂*r⁴)
Computes zenith angle θ from pixel radius r and focal length f, incorporating empirical fisheye correction terms for accurate sky coordinate registration.
Radiometric Uncertainty Propagation (GUM Type B)
u_c(Y) = √[∑(∂Y/∂x_i)² * u²(x_i) + 2*∑∑(∂Y/∂x_i)(∂Y/∂x_j)*u(x_i,x_j)]
Combines uncertainties from NIST-calibrated reference sources, detector nonlinearity, and temperature drift to determine expanded uncertainty (k=2) for digital number to irradiance conversion.