Inhaltszusammenfassung

The global photovoltaic market has developed very fast in the last years. In order to help PV system operators to decide whether their PV system is running well or not, we present a data cleansing algorithm to evaluate the specific energy yield of photovoltaic (PV) systems. The aim of our algorithm is to identify outlying yield and insolation values by using statistical methods and to provide a normal distribution of yield data after the final step. Therefore, PV systems are separated in majo...The global photovoltaic market has developed very fast in the last years. In order to help PV system operators to decide whether their PV system is running well or not, we present a data cleansing algorithm to evaluate the specific energy yield of photovoltaic (PV) systems. The aim of our algorithm is to identify outlying yield and insolation values by using statistical methods and to provide a normal distribution of yield data after the final step. Therefore, PV systems are separated in major and subregions. The specific energy yield and insolation values of all systems in a region are analyzed. Outlying values are neglected in further steps. The relevant key figure – the skew of the yield data distribution – converges within four passes of data cleansing steps and the confidence interval of the specific yield reaches 95 % if at least 50 PV systems are evaluable in one region» weiterlesen» einklappen

Autoren

Klassifikation

DDC Sachgruppe:
Allgemeines, Wissenschaft