Effects of Different Data Collection Strategies on Outdoor Performance Measurements and Degradation

Abstract

The accuracy of any analysis of PV module characteristics measured outdoor is normally affected by the following factors: meteorological conditions experienced during the measurements, locations where module are installed, different instrumentation used for measuring and recording approaches. Results can also be changed by the strategies used for data selection and analysis. With the aim of investigating degradation in modules, the performance parameters require a narrow uncertainty are needed as typically degradation rate of modules is less than 1% per year. A data collection and analysis strategy is presented that allow such an accuracy for PV modules deployed outdoors. This is then used for the analysis of long-term measurement campaign of amorphous silicon devices. The proposed data selection approach uses ISC as a measure of irradiance, which is also known as self-reference. This minimizes the spectral effects as well as the angle of incidence effects. Probability density functions (pdf) are used, which provide more information for the analysis of module degradation, i.e. the distribution of parameters may reflect the degradation modes of modules. The pdf results of PMPP, FF and VOC are presented and compared to the results of averages +/standard deviations. Distributions shift slightly over year indicating the degradation in modules. The uncertainties of performance parameters could be reduced to less than 1% for yearly results with a confidence level above 60% for the modules operating at CREST.