Uncertainty Considerations in Bifacial Photovoltaic Systems with High Albedo Seasonality

Bifacial photovoltaic systems are experiencing an increased deployment with the promise of lower generation costs. The energy gain of these systems highly depends on the albedo, which is uncertain and changes over time. This study examines the deviations in the energy yield predictions as a function of albedo at different temporal resolution from local, nearby and satellite datasets. DNV's SolarFarmer is used to model a three-year public dataset of a bifacial horizontal single-axis tracking PV system and validate the plane-of-array irradiance and DC energy yield modelling. The results showed that sub-hourly calculation timesteps can provide more accurate predictions. Similarly, high temporal resolution of albedo can benefit intra-daily energy yield predictions, but daily and monthly albedo can be sufficient for accumulated energy yield over longer periods of time. Accuracy differences were found between albedo source data, where satellite-derived albedo reported up to 2.1 % error increase compared to energy calculations using local site or nearby albedo data. The findings of the paper provide insights about the spatiotemporal resolution of albedo and its effects for energy assessment.