The purpose of this study is to conduct data mining research on building time series energy consumption data set. Research done so far mainly focused on cluster analysis. In this research, python language is used as the carrier, K-shape algorithm is adopted to perform cluster analysis on the time series energy consumption data from the perspective of time series. By analyzing the characteristics of its time series changes, the corresponding energy consumption patterns are obtained, and then corresponding energy saving measures are proposed to complete the construction of the entire method system. The final results show that the energy consumption curve obtained by clustering algorithm can effectively reflect the operation characteristics of the building. At the same time, based on the principle of peak cutting and valley filling, the energy consumption curve can be adjusted according to the characteristics of different modes, so that the building can effectively achieve the effect of energy saving.
Keywords Cluster analysis, Energy consumption data, Time series, Energy profile