SAEBKA: Similarity Analysis of Electricity Behavior based on K-shape Algorithm in AMI System

As China continues to strengthen its focus on the “double carbon” development goal, the power industry urgently needs to promote industrial restructuring in order to improve customer service quality. In the new generation of electricity information collection system, clustering the electricity data on AMI and finding the similarity patterns is the key to uncovering the features of customer behavior and improving the quality of power service. However, electricity data are time-series in nature, and it is difficult to get better results by traditional analysis methods. To address this problem, this paper proposes a K-shape based similarity analysis method for AMI electricity consumption behavior, which can realize fast feature mining and fine classification of electricity consumption behavior. The results show that it has obvious effect in enhancing the efficiency of electricity data analysis and improving the quality of customer service.