Analysis of Plane-of-Array Irradiance Sensitivity to Solar Resource Database Biases

Solar resource databases used for estimating the plane-of-array irradiance for small- and medium-scale distributed photovoltaic (PV) systems are typically selected exclusively based on global horizontal irradiance bias. Since the databases are usually derived from satellite-based products and complemented by reanalysis models, this criterion is not always adequate because the biases associated with direct normal irradiance and diffuse horizontal irradiance may be amplified by the transposition model. This study investigates the bias effect through an analytical approach based on an isotropic transposition model and proposes practical guidelines for selecting databases to minimize errors when estimating the plane-of-array irradiance. The analytical approach, which was validated through numerical simulations parameterized with anisotropic models, demonstrates that the plane-of-array estimation errors depend primarily on latitude, the PV modules’ tilt and azimuth, as well as on the time of the year. The results of a case study conducted to validate the theoretical analysis under real-world conditions show that the method consistently selects the database with the lowest plane-of-array irradiation error.