摘要
电能计量是电力市场改革的技术基础,随着我国电力现货市场的逐步展开,市场交易对用户侧的计量条件提出了更高要求。为解决现货市场环境下用户分时电量数据缺失的问题,引入压缩感知理论,通过研究其矩阵稀疏变换方法以及重构算法对电量时间序列进行压缩重构,提出了一种基于k-means聚类分析的压缩感知电量数据修复方法。应用该方法对某冶炼厂的电量数据进行修复,并将其与传统的修复方法进行比较,结果显示该模型在电表读数连续缺失时具有良好的修复效果。
Electric energy metering is the technical basis of electric power market reform.With the gradual development of spot market of electric power in China,market transactions put forward higher requirements on measurement conditions of user side.In order to solve the problem of the lack of time-sharing electricity data of users in the spot market environment,this paper introduces the theory of compressed sensing,studies the matrix sparse transformation method and reconstruction algorithm to compress and reconstruct the time series of electricity,and proposes a repair method of compressed sensing electricity data based on k-means clustering analysis.The model is applied to repair the electricity data of a smelter and compared with the traditional method.The results show that the model has a good repair effect when the electricity meter reading is missing continuously.
作者
王睿琛
卢少平
段沛恒
应黎明
卜琪
王霞
WANG Ruichen;LU Shaoping;DUAN Peiheng;YING Liming;PU Qi;WANG Xia(Kunming Electric Power Trading Center,Kunming 650041,China;School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China)
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2021年第3期247-254,共8页
Engineering Journal of Wuhan University
基金
昆明电力交易中心有限责任公司管理咨询项目(编号:ZD65072-DJF19110263)。
关键词
压缩感知
数据缺失
聚类分析
数据修复
compression perception
data missing
clustering analysis
data recovery
