A shape‐based clustering algorithm and its application to load data

The popularity of smart metres has brought a huge amount demand-side data, which provides important information for the demand response power sector, to guide practitioners understand customers' electricity usage behaviours and patterns. Clustering analysis daily load data is an tool mining users' consumption habits achieve non-fixed market segmentation. Since time series, it inappropriate perform clustering directly without extracting targeted features. Therefore, according shape features curve, shape-based algorithm called BDKM proposed. first uses B-splines regression fit series extract morphological features, then objects are segmented based on dynamic warping distance by clustering. Finally, real world used prove effectiveness proposed regression.