An statistical model for the short-term albedo estimation applied to PV bifacial modules

Abstract

Albedo estimation is an essential input parameter for bifacial PV modules. However, there was no clear agreement on which albedo values should be used and how they should be measured, either with satellite measurements or using onsite measurements which can be obtained from nearby meteorological stations. Since long-term measurements are not always available for new PV system locations, short-term albedo measurements are also used as input parameters in PV system performance estimation models. Short-term albedo presents high variability due to factors such as weather, seasonality or changes in the surrounding surface among others. In addition, apparently random albedo variations of 60% can be observed, even during consecutive days or within the same day. Therefore, this study presents a two-parameter exponential model that modelates the albedo data statistical distribution with an error of less than 5% in all cases and from which it is possible to determine the reliability of the obtained data. This model has been evaluated at a set of locations, with different surface types and climates, and assessed using onsite and satellite data. The impact of the methodology used to estimate the albedo data uncertainty and its reliability has also been studied as a function of several parameters such as the global horizontal irradiance and measurement time, allowing the process optimization and enhancing its reliability.