电力负荷数据典型特征提取  

Typical Feature Extraction of Power Load Data

作　　者：周自强 赵淳[1,2] 范鹏 ZHOU Ziqiang;ZHAO Chun;FAN Peng(NARI Group Co.,Ltd.,Nanjing 211106,China;Wuhan NARI Limited Liability Company,State Grid Electric Power Research Institute,Wuhan 430074,China)

机构地区：[1]南瑞集团有限公司,江苏南京211106 [2]国网电力科学研究院武汉南瑞有限责任公司,湖北武汉430074

出　　处：《电工技术》2020年第21期69-71,共3页Electric Engineering

摘　　要：随着智能电网的发展,用户类别日趋多样化,电力用户行为特征的研究显得尤为重要。首先对原始数据进行降维和分解,接着进行聚类分析,提取典型用户的日负荷曲线,最后对某地区不同用户的电力负荷曲线进行试点和应用。研究结果表明,该聚类分析方法准确率高,可提取出四类典型负荷曲线,为电力公司提供个性化、精准性的售电服务提供了理论依据。With the rapid development of smart grid and the diversification of user types,the research on the behavior characteristics of power users is particularly important.In this paper,firstly,the original data was degraded and decomposed.Then,the clustering analysis was carried out to extract the daily load curve of typical users.Finally,the power load curve of different users in a certain region was piloted and applied.The research results showed that the cluster analysis method had high accuracy and can extract 4 groups of typical load curves.The research results provided a theoretical basis for power companies to provide personalized and accurate electricity sales services.

关 键 词：智能电网 SVD KL散度 电力负荷 特征提取 聚类分析 

分 类 号：TM715[电气工程—电力系统及自动化]