Rating of annual energy yield more sensitive to reference power than module technology

Abstract

At present, many different photovoltaic (PV) technologies share the market. Especially investors want to know how much energy each of the PV technologies produces. This paper discusses the measured annual energy yield E <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AC</sub> of twelve PV technologies under different climatic conditions in Germany and Cyprus over three years of operation. In order to compare the annual yield of different PV technologies, the E <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AC</sub> data are normalized to the rated power P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</sub> , to the flasher power P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">flash</sub> , and to the measured field power P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">field</sub> . An error analysis is done for both, the energy measurement E <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AC</sub> and the nominal power P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">STC</sub> . It is found that the typical uncertainty for an energy yield comparison is ±5 %. This means that a difference of 10 % in the annual energy yield between PV technologies can not be traced back to the technologies themselves. The performance analysis of all PV systems shows that the differences in the energy yield are smaller than the error bars on the reference power. Therefore it is not yet possible to decide which PV technology is the best. Moreover, despite obvious trends on the data, we can not unambiguously conclude that PV modules with a better temperature or low light behavior will ensure a higher energy yield in general, since the propagation of state-of-the-art nominal power rating errors outbalances the well recognized effects of low light and temperature dependencies.