Sky radiance distribution based model for rear and front insolation estimation on PV bifacial modules
Year: 2025
Type: article
Abstract: Single-axis bifacial photovoltaic tracking systems enable an energy production increase through the conversion of ground-reflected irradiance. This study presents a model to determine the irradiance distribution on both sides of bifacial module arrays to calculate best tilt and motion laws for maximizing irradiance collection at different sky conditions. The 3D celestial vault is described through the Perez “All-Weather” sky model, thus comprising different radiance distributions across the sky dome. The model relies on calculations of view factors related to sky portions (average solid angle 0.0376 Sr) and module and ground subareas; its results have been validated against simulations performed through “Bifacial_radiance” by NREL. The hourly annual analysis has been conducted both for a Mediterranean and for a continental plant to compare the solar yield from the present model tracking law to backtracking strategy based on sun position at different azimuthal orientations of the plant. Solar energy collection increase can reach 6 % during months characterized by a larger share of diffuse irradiance. Calculation of irradiance distribution on front and rear module surfaces allows for precise estimation of available solar energy and for the definition of strategies aimed at maximizing productivity while considering cell-to-cell mismatch and “bifaciality factor” effects in adjacent arrays. • Backtracking isn't the optimal strategy in all weather conditions. • Weather effect on sky radiance distribution has been assessed using a new model. • Model validity is assessed against raytracing software bifacial_radiance (R 2 = 0.99). • The improved tracking strategy allows monthly insolation gains up to 6 %. • Higher gains are possible for east-west oriented plants and cloudy periods.
Authors: Mattia Parenti, Samuele Memme, Marco Fossa
Institution: University of Genoa
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $4,340
OpenAlex ID: W4410079989