浙江电力

2017, v.36;No.260(12) 33-36+62

[打印本页] [关闭]
本期目录(Current Issue) | 过刊浏览(Archive) | 高级检索(Advanced Search)

基于曲线相似度和关联分析的窃电智能识别与预警
Intelligent Identification and Early Warning against Power Theft Based on Curve Similarity and Correlation Analysis

陈仕军,王长江
CHEN Shijun,WANG Changjiang

摘要(Abstract):

针对近年来传统生产加工类实体企业愈加突出的窃电问题,应用用电信息采集系统、营配贯通平台及营销业务应用系统数据,以配电房线损异常波动为主线,利用关联分析算法、结果判定区间和主题权值等建立智能诊断关联分析模型,分析表计开盖、表计装拆、用户电量波动和终端异常告警事件数据,实现对窃电及违约用电异常用户的精确定位。结合实例验证了该方法的有效性,并对后续研究进行了展望。
With power theft increases in traditional production and manufacturing enterprises in recent years,data from electricity information acquisition system, marketing and distribution integration platform and marketing business application system are used, and correlation analysis, result judgment interval and theme weights are applied to establish an intelligent diagnosis correlation analysis model with abnormal fluctuation of line loss as the mainline to analyze meter opening, meter installation and disassembly, power consumption fluctuation and terminal abnormity warning event data to precisely localize power theft and abnormal power users. The method is verified by some cases, and the prospect of future research is surveyed.

关键词(KeyWords): 窃电;智能识别;预警;关联分析算法;曲线相似度;结果判定区间;主题权值
power theft;intelligent identification recognition;warning;correlation analysis;curve similarity;result judgment interval;theme weight

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 陈仕军,王长江
CHEN Shijun,WANG Changjiang

DOI: 10.19585/j.zjdl.201712007

参考文献(References):

扩展功能
本文信息
服务与反馈
本文关键词相关文章
本文作者相关文章
中国知网
分享