Locational Performance Analysis of Distributed Photovoltaic Systems

Big Data impose new research challenges in power system industry. Volume and variety of Big Data are crucial parameters for the validation of data processing techniques. Due to the rapid augmentation of Photovoltaics (PVs) installations in many countries the volume of the collected on-site data continually increases. The scope of the present paper is to provide an analysis on an amount of PV power data for the purpose of decreasing the need for continuous measurements and therefore, lowering the volume of the exploitable Big Data in PV sites. Data collected from 10 PV installations in Norther Greece are used as a case study. The correlation between each installation is calculated exhibiting high values in geographically nearby areas. Thus, metering and storing of data can be reduced to sporadic and seasonal measurements and in selected installations. Profiling can be used as well in order to reduce the needs for measurements, by taking as a benchmark the most appropriate installation out of geographically close ones. The profiles can be updated once a month by taking into account temperature variation, PV overall apparatus efficiency and other factors.