INDUSTRIAL ELECTRICITY PORTRAIT BASED ON IMPROVED K-MEANS CLUSTERING ALGORITHM
X. Yang, Shuxin Liang, Huageng Tian, Y. Wu, Xueyan Cheng · IET conference proceedings. · 2021
With the deepening of big data research and the construction of distribution network automation, power big data is getting more and more attention. How to realize the perception of power network users based on the big data of distribution network and support the related business of distribution network is the key issue that power companies pay attention to at present. In this paper, the annual row data of users are pre-processed, and the electricity consumption data are clustered by improved k-means algorithm. Then combined with the user's industry information, through the Granger causality test and other algorithms, the industry electricity from the characteristics of electricity and power driving effect of the two dimensions of the portrait; Finally, based on the actual data, the industry electricity portrait is realized, and the effectiveness of the method is verified.