摘要
设备负荷监测是能源管理的关键,允许用户获得特定于设备的能源消耗统计数据,可有效用于设计负载调度策略。非侵入式负荷监测(Non-intrusive load monitoring,NILM)在能源分解方面效果显著,允许从单个测量点采集的聚合数据中识别设备类别和功耗。论文对在国内外开展的非侵入式负荷监测研究进行综述。首先凝练了NILM问题模型和研究框架;重点概述了关键技术,即事件检测、特征提取、负荷识别,深入讨论了关键技术目前研究路线和不同方法之间优劣点,针对研究人员为开发准确NILM系统以实现有效能源管理而探索的最新算法做了重点分析,讨论了NILM系统精准评估框架;最后对现有研究中存在的挑战剖析,展望面向新型电力系统框架的研究方向。
Equipment load monitoring is key to energy management,allowing users to obtain equipment-specific energy consumption statistics that can be effectively used to design load scheduling strategies.Non-intrusive load monitoring(NILM)is highly effective in energy disaggregation,allowing the identification of equipment categories and power consumption from aggregated data collected at individual measurement points.This paper presents a review of the research on non-intrusive load monitoring carried out in China and abroad.It first condenses the NILM problem model and research framework,highlights an overview of key technologies,namely event detection,feature extraction,and load identification,discusses in depth the current research lines of key technologies and the advantages and disadvantages between different approaches,provides a focused analysis of the latest algorithms explored by researchers to develop accurate NILM systems for effective energy management,and discusses a framework for accurate evaluation of NILM systems,and finally,challenges in existing research are profiled and research directions towards a new power system framework are foreseen.
作者
钱玉军
包永强
姜丹琪
张旭旭
雷家浩
QIAN Yujun;BAO Yongqiang;JIANG Danqi;ZHANG Xuxu;LEI Jiahao(Institute of Artificial Intelligence Industry Technology,Nanjing Institute of Technology,Nanjing 211167;School of Information and Communication Engineering,Nanjing Institute of Technology,Nanjing 211167)
出处
《计算机与数字工程》
2023年第2期380-389,共10页
Computer & Digital Engineering
基金
国家自然科学基金青年基金项目(编号:62171217)资助。
关键词
非侵入式负荷监测
事件检测
特征提取
负荷识别
新型电力系统
non-intrusive load monitoring
event detection
feature extraction
load identification
new power system
