Simulation of Large PV Plants Using a Continuous Radiance Distribution Model and Cell-Resolution Mismatch Calculation

Kurzfassung

The main goal of the present work is to study the sensitivity of shading calculations under a "realistic" (quasi-continuous) radiance distribution model, as opposed to the "Standard" isotropic assumption. We are interested in the balance between the possible accuracy improvements vs the increased computational expense, assuming the shading calculation is coupled with an accurate electrical mismatch model and applied to large-scale PV systems. Two existing PV plants are modeled under three scenarios: (a) an isotropic 3D view-factor (b) a three-component model (with a horizon-brightening band); and (c) a continuous radiance distribution function modeled as a detailed (43 element) sky-tessellation. To estimate a metric of the sensitivity of the model, a cross comparison of numerical results for a small set of virtual PV plants with diverse configurations and climatic conditions is performed. Finally, indirect validation of the model is attempted by comparison of the simulation output to plant monitoring data, but the data quality and resolution turn out to be completely insufficient to reflect the subtlety of the diffuse shading effect.