异常用电的检测方法：评述与展望

Detection Methods of Abnormal Electricity Consumption Behaviors:Review and Prospect

用电侧的非技术性损失(NTL)主要由用户异常用电行为造成,一直是各国电网企业期望解决的重要问题。随着智能电表在电网中的普及,利用配电网、用户侧的量测数据实现对于NTL的高效、精准检测,受到了学术界和业界的广泛关注。首先介绍用户异常用电行为的基本模型、方法与评价指标。随后,从基于系统状态、基于数据驱动和基于博弈论三个角度对现有的异常用电行为检测方法进行系统性的梳理、分析与比较,以总结出其对于后续研究工作开展的借鉴意义。最后,对异常用电行为检测领域的关键技术进行了提炼,并对异常用电行为检测领域未来的研究工作进行了展望。

Non-technical loss（NTL）on customer side is mainly caused by users＇ abnormal electricity consumption behaviors and has always been an important issue that power utilities from various countries aim to solve.It also receives wide concerns from independent electricity distribution and retail companies in the deregulated power market.With the rising penetration of smart meters in the power system,the problem of utilizing measurement data in the distribution network and customer side to detect NTL efficiently and precisely has received considerable attention worldwide.Firstly,this paper introduces the basic model of abnormal electricity consumption behaviors and the evaluation criteria of the detection methods.The present detection methods are systematically analyzed and compared from the aspects of system-state based,data-driven based and game-theory based,to summarize their referential value for the future research.Finally,the key technologies in NTL detection are extracted,and some future research orientations in the field of abnormal electricity consumption behaviors are discussed.