龙源期刊网 http://www.qikan.com.cn
基于大数据的用电行为异常分析
作者:彭涛
来源:《中国科技纵横》2017年第05期
摘 要:智能电网是大数据的重要技术应用领域之一。随着智能电网的发展, 窃电犯罪日渐网络化、规模化、专业化。由于高级量测体系、各种监控系统的大规模部署产生和积累了大量数据, 通过信息采集并充分挖掘这些数据的价值具有重要意义。针对智能配用电业务,首先分析智能配用电大数据的特征,然后重点研究大数据环境下的用户用电行为应用场景,提出大数据环境下的用户用电行为异常研究思路和方法,接着分析业务应用中的大数据关键技术。 关键词:大数据;用电行为;异常分析
中图分类号:TP319 文献标识码:A 文章编号:1671-2064(2017)05-0158-04 Abstract: Smart grid is one of the most important technical fields for big data technology application. With the development of smart grid, Stealing is growing crime network, scale and specialization. Since the deployments of advanced metering infrastructure(AMI), equipment condition monitoring systems result in the production and accumulation of a lot of data, thus it is of great significance to fully mine the value of these data and data collection. Firstly, aiming at smart power distribution and consumption systems, the paper described the big data and its characteristics. Secondly, the typical application scenario analyses was carried out, which were customer
electricity usage behavior analysis. Then the research methods of the business application in big data environment were put forward. Finally, necessary big data key technologies were proposed on big data in power distribution and consumption systems was presented.
Key words:Big data; electricity usage behavior; abnormal Analysis 1 引言
近年来,盗窃电能的违法行为越来越普遍,而且随着科学技术的快速发展,窃电手段更加高明,技术含量更高,方法更隐蔽,特别是窃电犯罪日渐呈现网络化、规模化、专业化的特点,给查处带来了很大的难度。窃电现象不仅困扰电力企业的发展,对电网安全与效益、社会安全与风气构成了严重挑战,也严重影响了国家的经济建设和社会的稳定。
尽管窃电的形式和手段很多,无论是哪一种窃电方式,都会影响某个电表的计量数据或者影响某条线路、某个区域的相关数据,如线损变化、电压变化、电流变化以及相关的电表事件等。随着用电信息采集范围、应用范围的持续扩大,伴随着用户量和业务需求的增长,系统的复杂度也大幅度增加。这些数据在当今互联网背景下愈发体现了大数据的关键要素[1]:容量(Volume),速度(Velocity),种类(Variety),不仅体量巨大,数据类型繁多,价值密度低且商业价值高。传统数据处理技术已无法满足当前业务发展对系统数据处理分析的要求。通