Modelling the Performance of Near-Ultraviolet Organic Solar Cells Under Meteorologically Representative Spectral Irradiances

Abstract

The introduction of near-UV solar cells provides great potential for the proliferation the self-powered electrochromic window (ECW). The viability of such a pairing requires extensive performance testing. Traditionally, the solar industry artificially simulates ideal testing conditions for use on well-characterized crystalline silicon cells. However, new PV prototypes generally face limited testing outside the lab, and standardized irradiance testing schemes fail to paint a holistic picture of how they perform in real-world settings. This thesis project aims to address this discrepancy by leveraging cluster analysis on publicly available solar irradiance data from the National Solar Radiation Database. Its goals are three-fold:

1.) Develop representative spectral irradiances for each of 9 climatically consistent regions within the contiguous United States. 2.) Model the energy production of north, south, east, and west-facing near-UV solar cells under the conditions developed for each region. 3.) Develop granular models that account for obstructions in the urban context.

In summary, the use of of a k-means clustering algorithm was sufficient in creating resolved clusters that captured seasonal differences in the magnitude of modeled irradiance. In addition, all regions and orientations demonstrated sufficient productivity over a 7-hour period to power a typical electrochromic window (ECW) for 24 hours. Even in the least productive case, 8 times more energy was produced than was required. Nevertheless, more granular investigations emphasize the importance of the local layout, as urban obstructions were shown to significantly stunt power output.