PHOTOVOLTAIC SYSTEM YIELD EVALUATION IN SWEDEN : A performance review of PV systems in Sweden 2017-2018

The goal of this study is to evaluate Swedish photovoltaic systems regarding energy production from two different years and compare the gathered data with results from a model simulating optimal conditions. This is done to investigate how the energy production differs between each year, why there are differences, and also to evaluate the simulation tools compared to the real production data. A good way to measure performance is to calculate the specific yield, that is the energy produced per unit of installed power (kWh/kWp). In order to complete this study, a literature study was made to investigate reasons for potential variations in PV system yield. Besides that, the production data from 2373 PV systems in Sweden were collected from different databases, and the data were sorted and compiled in order to calculate specific yield (kWh/kWp). The total number of PV systems after sorting was 828 for the 2017-2018 data and 1380 systems for the 2018 data. Data from real PV system production was compared with calculations performed in two simulation tools, PVGIS and PVsyst. Differences in calculation methods were investigated for performance evaluations between the two programs, and also for comparison with the real plant data. The results showed that the average specific yield for Sweden as a whole, to be 798 kWh/kWp for 2017. For 2018 with the results where 890 kWh/kWp when looking at the exact same plants as for 2017. This is an increase of 11,5%. For the simulation tools the results where 974 kWh/kWp for PVGIS, and 978 for PVsyst for an optimized system. Larger variations in specific yield occurs between every of the 21 counties in Sweden. The solar irradiations show significant correlations to the variations of the 2017 and 2018 specific yield data. Differences between the production data from the two years and the simulation tools wereinvestigated further. Reasons for this was discussed to be because of orientations of the panels and shading of the panels. Real PV systemsdiffer in orientation and the amount of shadowing from the simulated calculations.