Photovoltaic power pattern grouping based on bat bio-inspired clustering

Photovoltaic power pattern (PVPP) clustering is a key tool to provide information about the impacts of the interconnection of photovoltaic systems onto the electric distribution system without extensive analysis and simulations. This paper presents a bio-inspired clustering method to group PVPPs. The Bat clustering method is illustrated, highlighting its characteristics and parameters, during the iterative process. Furthermore, the results of Bat clustering method are compared with those obtained from the classical K-means and Ward's Hierarchical clustering methods using three internal validity indices. The clustering results show that Bat was the most efficient method and outperformed the other clustering methods.