Cross-characterization of PV and sunshine profiles based on hierarchical classification

This paper investigates the link between PV generation and the climatological indicator of sunshine duration. For the first time, an understanding of to what degree the daily sunshine duration determines the generation output profile is established, and insights into the extent of such impact at differing months of a year are provided. The new finding essentially provides a fresh new perspective on characterizing the uncertainty and variability in PV output. Based on this correlation identified, a novel two-step hierarchical classification method is also proposed in this work to facilitate PV profiling. A case study on a practical PV plant in Great Britain is presented to demonstrate the application of this method. For each derived group, the degree of variation in PV output at different times and the confident levels of each quantity are assessed. More importantly, based on the classification results, a weather-based PV profiling guideline is created. This will facilitate PV output forecasting on a granular level, thus providing a powerful tool for the ever increasingly challenging system operation and planning.