Topology Verification of Low Voltage Distribution Network Based on k-means Clustering Algorithm

2019 2nd International Conference on Advanced Materials, Intelligent Manufacturing and Automation

Wang, Junyi^1 ; Ji, Xingquan^1 ; Li, Kejun^2 ; Sun, Qiaoyu^1

School of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, Shandong

266590, China^1

School of Electrical Engineering, Shandong University, Jinan, Shandong

250061, China^2

关键词: Automatic corrections;    Improved k-means clustering;    Low voltage distribution network;    Low voltage substations;    Measurement system;    Noise processing;    Topology verification;    Transformer stations;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/569/5/052096/pdf
DOI  :  10.1088/1757-899X/569/5/052096

来源: IOP

【 摘 要 】

Aiming at the problem of topology connection error existing in GIS system of low voltage distribution network, a topology verification algorithm using AMI voltage measurement data combined with k-means clustering algorithm is proposed. Firstly, the voltage data of the consumers in the low-voltage substation area is obtained by the AMI measurement system. Then k-means clustering algorithm is used by the similarity to cluster the voltage curves to identify and verify the users connected on the incorrectly transformer stations. An improved method of noise processing using data density set is proposed to solve the problem of initial cluster centre selection in k-means. For the problem of k value selection, an automatic correction of optimal k value is proposed. The feasibility and effectiveness of the improved k-means clustering algorithm in topology verification are verified by a practical example.