On the methodology of energy yield assessment for one-Sun tandem solar cells

In this paper we compare different energy yield calculation methods for non-concentrating (i.e., 1-Sun) tandem solar cells, which are believed to be a viable next-generation high-efficiency photovoltaic (PV) concept. The yield calculation methods use illumination inputs with different levels of detail and accuracy. Through this exercise, we show how subtleties in the temporal resolution and accuracy of the illumination input affect the calculated energy yield in the theoretical analysis of the expected outdoor performance of 1-Sun tandem solar cells. The algorithm which we use to compute the energy yield is computationally efficient and is based on the average photon energy of a given optical spectrum. This approach involves device simulation, which is particularly relevant for solar cells and modules at the research and development stages. Energy yield calculations using this approach are performed for different time scales and are compared to results from a reference calculation. For short-term yield calculations, a detailed input of illumination conditions with high temporal resolution is necessary for obtaining accurate results. For long-term yield calculations, it is less important to capture rapid fluctuations. It is found that using simulated spectra to evaluate spectral effects, as is routinely done for concentrating PV devices, is not accurate enough for 1-Sun tandem solar cells, as it can underestimate the losses in the performance ratio by as much as 60%. Our analysis also indicates that the availability of detailed and realistic illumination conditions is important for the design of 1-Sun tandem solar cells.