基于k-medoids聚类算法的异常低压台区线损识别方法研究  

Research on Line Loss Identification Method of Abnormal Low-Voltage Station Area Based on k-medoids Clustering Algorithm

作　　者：吴化委 强利军 马晓伟 Wu Huawei;Qiang Lijun;Ma Xiaowei(State Grid Xinjiang Power Co.,Ltd,Kuitun Power Supply Company,Yili Xinjiang 833200,China)

机构地区：[1]国网新疆电力有限公司奎屯供电公司,新疆伊犁833200

出　　处：《现代工业经济和信息化》2024年第11期286-288,共3页Modern Industrial Economy and Informationization

摘　　要：研究通过引入k-means聚类算法,致力于在复杂电网环境中提高线损准确性。通过采集大量电力系统数据,运用k-medoids聚类算法对数据进行分组,识别异常模式。研究结果表明,k-medoids聚类算法在异常低压台区线损识别方面表现出色,相较于传统方法更具准确性和鲁棒性。在实际应用中,该方法成功应用于特定电网台区,实现了对异常线损的实时监测和精准定位。本次研究对于提高电力系统稳定性以及工作效率提供了技术支持,促进了电力行业的可持续发展。The study is aimed at improving the accuracy of line loss in complex grid environment by introducing clustering algorithm.A large amount of power system data is collected,and the clustering algorithm is applied to group the data to identify abnormal patterns.The results show that the clustering algorithm performs well in the identification of abnormal low-voltage areas,and is more accurate and robust than the traditional methods.In practical application,the method is successfully applied to specific grid areas,realizing real-time monitoring and accurate localization of abnormal line loss.This study provides technical support to improve the stability and efficiency of the power system,and promotes the sustainable development of the power industry.

关 键 词：k-medoids聚类算法 异常低压台区 线损 

分 类 号：TM615[电气工程—电力系统及自动化]